pip install pandas

Requirement already satisfied: pandas in ./myenv3.10/lib/python3.10/site-packages (2.3.0)
Requirement already satisfied: tzdata>=2022.7 in ./myenv3.10/lib/python3.10/site-packages (from pandas) (2025.2)
Requirement already satisfied: numpy>=1.22.4 in ./myenv3.10/lib/python3.10/site-packages (from pandas) (1.26.4)
Requirement already satisfied: pytz>=2020.1 in ./myenv3.10/lib/python3.10/site-packages (from pandas) (2025.2)
Requirement already satisfied: python-dateutil>=2.8.2 in ./myenv3.10/lib/python3.10/site-packages (from pandas) (2.9.0.post0)
Requirement already satisfied: six>=1.5 in ./myenv3.10/lib/python3.10/site-packages (from python-dateutil>=2.8.2->pandas) (1.17.0)

[notice] A new release of pip available: 22.2.2 -> 25.1.1
[notice] To update, run: pip install --upgrade pip
Note: you may need to restart the kernel to use updated packages.

pip install numpy==1.26.4

Requirement already satisfied: numpy==1.26.4 in ./myenv3.10/lib/python3.10/site-packages (1.26.4)

[notice] A new release of pip available: 22.2.2 -> 25.1.1
[notice] To update, run: pip install --upgrade pip
Note: you may need to restart the kernel to use updated packages.

pd.set_option('display.max_columns', None)

pip install nfl_data_py

Requirement already satisfied: nfl_data_py in ./myenv3.10/lib/python3.10/site-packages (0.3.3)
Collecting pandas<2.0,>=1.0
  Using cached pandas-1.5.3-cp310-cp310-macosx_10_9_x86_64.whl (12.0 MB)
Requirement already satisfied: appdirs>1 in ./myenv3.10/lib/python3.10/site-packages (from nfl_data_py) (1.4.4)
Requirement already satisfied: numpy<2.0,>=1.0 in ./myenv3.10/lib/python3.10/site-packages (from nfl_data_py) (1.26.4)
Requirement already satisfied: fastparquet>0.5 in ./myenv3.10/lib/python3.10/site-packages (from nfl_data_py) (2024.11.0)
Requirement already satisfied: packaging in ./myenv3.10/lib/python3.10/site-packages (from fastparquet>0.5->nfl_data_py) (25.0)
Requirement already satisfied: cramjam>=2.3 in ./myenv3.10/lib/python3.10/site-packages (from fastparquet>0.5->nfl_data_py) (2.10.0)
Requirement already satisfied: fsspec in ./myenv3.10/lib/python3.10/site-packages (from fastparquet>0.5->nfl_data_py) (2025.5.1)
Requirement already satisfied: python-dateutil>=2.8.1 in ./myenv3.10/lib/python3.10/site-packages (from pandas<2.0,>=1.0->nfl_data_py) (2.9.0.post0)
Requirement already satisfied: pytz>=2020.1 in ./myenv3.10/lib/python3.10/site-packages (from pandas<2.0,>=1.0->nfl_data_py) (2025.2)
Requirement already satisfied: six>=1.5 in ./myenv3.10/lib/python3.10/site-packages (from python-dateutil>=2.8.1->pandas<2.0,>=1.0->nfl_data_py) (1.17.0)
Installing collected packages: pandas
  Attempting uninstall: pandas
    Found existing installation: pandas 2.3.0
    Uninstalling pandas-2.3.0:
      Successfully uninstalled pandas-2.3.0
Successfully installed pandas-1.5.3

[notice] A new release of pip available: 22.2.2 -> 25.1.1
[notice] To update, run: pip install --upgrade pip
Note: you may need to restart the kernel to use updated packages.

import nfl_data_py as nfl
import pandas as pd
import numpy as np
weekly_stats = nfl.import_weekly_data(years = range(2010, 2025))
weekly_stats.rename(columns={'recent_team':'team'}, inplace=True)
weekly_stats

Downcasting floats.

# Importing schedule data
schedules = nfl.import_schedules(years = range(2010, 2025))
schedules['time'] = pd.to_datetime(schedules['gametime'])
schedules['hour'] = schedules['time'].dt.hour
schedules['slate'] = np.where(
    schedules['hour'].isin([12, 13, 14]),
    'Morning',
    np.where(
        schedules['hour'].isin([15]),
        'Noon',
        np.where(
            schedules['hour'].isin([16, 17, 18, 19]),
            'Afternoon',
            np.where(
                schedules['hour'].isin([20, 21, 22, 23]),
                'Night',
                'global'
            )
        )
    )
)

# Unifying abbreviations of teams that have moved (and thus changed their abbrevaitions)
team_map = {'STL':'LA', 'SD':'LAC', 'OAK':'LV'}
schedules['home_team'] = schedules['home_team'].replace(team_map)
schedules['away_team'] = schedules['away_team'].replace(team_map)

# Creating day_slate column
schedules['day_slate'] = schedules['weekday'] + '_' + schedules['slate']

# Creating location column
schedules['location'] = np.where(
    schedules['stadium'].isin(['Wembley Stadium', 'Rogers Centre', 'Twickenham Stadium', 'Azteca Stadium', 'Tottenham Stadium',
                                  'Allianz Arena', 'Deutsche Bank Park', 'Arena Corinthians']), 'global', schedules['home_team'])

schedules

# Merging weekly_stats and schedules
weekly_stats['matchup'] = weekly_stats.apply(lambda row: '_'.join(sorted([row['team'], row['opponent_team']])), axis=1)
schedules['matchup'] = schedules.apply(lambda row: '_'.join(sorted([row['home_team'], row['away_team']])), axis=1)
weekly_stats = weekly_stats.merge(schedules[['season', 'week', 'matchup', 'game_type', 'day_slate', 'location']], how='left', on=['season', 'week', 'matchup'])
weekly_stats

# Creating game_number column
game_no = weekly_stats[['season', 'week', 'team', 'location']].groupby(['season', 'team', 'week']).agg({'location': 'first'})
game_no['game_number'] = game_no.groupby(['season', 'team']).cumcount() + 1
game_no.reset_index(inplace=True)
weekly_stats = weekly_stats.merge(game_no, how='left', on=['season', 'team', 'week', 'location'])
weekly_stats

# Reformatting rows and columns
weekly_stats = weekly_stats.sort_values(['season', 'week'])
cols = list(weekly_stats.columns)
weekly_stats = weekly_stats[cols[:3] + [cols[-1]] + cols[3:-1]]
weekly_stats.reset_index(inplace=True, drop=True)
weekly_stats

# Limiting to fantasy weeks only
weekly_stats = weekly_stats[
    ((weekly_stats['season'] <= 2020) & (weekly_stats['week'] <= 16)) | 
    ((weekly_stats['season'] >= 2021) & (weekly_stats['week'] <= 17))
]

# Limiting to relevant positions only
weekly_stats = weekly_stats[weekly_stats['position_group'].isin(['WR', 'QB', 'TE', 'FB', 'RB'])]

# Dropping unnecessary columns
weekly_stats = weekly_stats.drop(columns={
    'position_group', 'headshot_url', 'player_name', 'season_type', 'matchup', 'game_type'
})
weekly_stats.reset_index(inplace=True, drop=True)
weekly_stats

# Define meta/stat columns
meta_cols = ['player_id', 'player_display_name', 'position', 'team', 'season', 'week', 'game_number', 'opponent_team']
stat_cols = [col for col in weekly_stats.columns if col not in meta_cols + ['day_slate', 'location']]

# Copy original dataframe
current_df = weekly_stats.copy()
prior_df = weekly_stats.copy()
prior_df['season'] += 1  # Offset to match with current season

# Get average stats vs each opponent in the prior season
opponent_avg = (
    prior_df
    .groupby(['player_id', 'season', 'opponent_team'])[stat_cols]
    .mean()
    .reset_index()
    .rename(columns={col: f"{col}_lag_opponent" for col in stat_cols})
)

# Merge opponent-based lag stats into current_df
merged = current_df.merge(
    opponent_avg,
    how='left',
    left_on=['player_id', 'season', 'opponent_team'],
    right_on=['player_id', 'season', 'opponent_team']
)

# Identify rows with missing opponent lag values
missing_opponent = merged[[f"{col}_lag_opponent" for col in stat_cols]].isna().all(axis=1)

# Prepare game_number-based fallback from prior_df
fallback_avg = (
    prior_df
    .groupby(['player_id', 'season', 'game_number'])[stat_cols]
    .mean()
    .reset_index()
    .rename(columns={col: f"{col}_lag_game" for col in stat_cols})
)

# Merge fallback lag stats only to missing rows
fallback_df = current_df.loc[missing_opponent, ['player_id', 'season', 'game_number']]
fallback_df = fallback_df.merge(
    fallback_avg,
    how='left',
    on=['player_id', 'season', 'game_number']
)

# Merge fallback_df back into merged DataFrame
for col in stat_cols:
    lag_col = f"{col}_lag_opponent"
    fallback_col = f"{col}_lag_game"
    merged.loc[missing_opponent, lag_col] = fallback_df[fallback_col].values

# Add integer indicators for the source of the lag
merged['used_opponent_avg'] = (~missing_opponent).astype(int)
merged['used_game_number_avg'] = missing_opponent.astype(int)

# Identify rows with no lagged data (DNPs)
merged['DNP_prior_season_game'] = merged[f"{stat_cols[0]}_lag_opponent"].isna().astype(int)

# Fill with player prior season averages where DNP occurred
prior_season_avg = (
    weekly_stats
    .groupby(['player_id', 'season'])[stat_cols]
    .mean()
    .reset_index()
    .rename(columns={col: f"{col}_lag_avg_season" for col in stat_cols})
)

season_avg_df = current_df.loc[merged['DNP_prior_season_game'] == 1, ['player_id', 'season']].copy()
season_avg_df['season'] -= 1  # Adjust to prior season
season_avg_df = season_avg_df.merge(
    prior_season_avg,
    how='left',
    on=['player_id', 'season']
)

# Fill DNP rows with prior season averages
for col in stat_cols:
    merged.loc[merged['DNP_prior_season_game'] == 1, f"{col}_lag_opponent"] = season_avg_df[f"{col}_lag_avg_season"].values

# Identify "never_seen" players — no prior season record
seen_players = weekly_stats[['player_id', 'season']].drop_duplicates()
seen_players['season'] += 1
merged = merged.merge(
    seen_players.assign(seen_before=1),
    on=['player_id', 'season'],
    how='left'
)
merged['never_seen'] = merged['seen_before'].isna().astype(int)
merged.drop(columns='seen_before', inplace=True)

# Final lagged dataframe
lagged_stats = merged.copy()

# Deleting extra columns
cols_to_drop = list(lagged_stats.loc[:, 'completions':'fantasy_points'].columns) + ['fantasy_points_lag_opponent']
lagged_stats = lagged_stats.drop(columns=cols_to_drop)
lagged_stats.rename(columns=lambda col: col.replace('lag_opponent', 'lagged'), inplace=True)
lagged_stats.rename(columns={'player_display_name': 'player_name'}, inplace=True)
lagged_stats.loc[lagged_stats['position'] == 'HB', 'position'] = 'RB'

# Getting rid of null game_number columns - wrongly input data
lagged_stats = lagged_stats[lagged_stats['game_number'].notna()]
lagged_stats['season'] = lagged_stats['season'].astype(str)
lagged_stats['week'] = lagged_stats['week'].astype(str)
lagged_stats['game_number'] = lagged_stats['game_number'].astype(str)
lagged_stats

pip install seaborn

Requirement already satisfied: seaborn in ./myenv3.10/lib/python3.10/site-packages (0.13.2)
Requirement already satisfied: matplotlib!=3.6.1,>=3.4 in ./myenv3.10/lib/python3.10/site-packages (from seaborn) (3.10.3)
Requirement already satisfied: numpy!=1.24.0,>=1.20 in ./myenv3.10/lib/python3.10/site-packages (from seaborn) (1.26.4)
Requirement already satisfied: pandas>=1.2 in ./myenv3.10/lib/python3.10/site-packages (from seaborn) (1.5.3)
Requirement already satisfied: packaging>=20.0 in ./myenv3.10/lib/python3.10/site-packages (from matplotlib!=3.6.1,>=3.4->seaborn) (25.0)
Requirement already satisfied: kiwisolver>=1.3.1 in ./myenv3.10/lib/python3.10/site-packages (from matplotlib!=3.6.1,>=3.4->seaborn) (1.4.8)
Requirement already satisfied: pyparsing>=2.3.1 in ./myenv3.10/lib/python3.10/site-packages (from matplotlib!=3.6.1,>=3.4->seaborn) (3.2.3)
Requirement already satisfied: contourpy>=1.0.1 in ./myenv3.10/lib/python3.10/site-packages (from matplotlib!=3.6.1,>=3.4->seaborn) (1.3.2)
Requirement already satisfied: cycler>=0.10 in ./myenv3.10/lib/python3.10/site-packages (from matplotlib!=3.6.1,>=3.4->seaborn) (0.12.1)
Requirement already satisfied: pillow>=8 in ./myenv3.10/lib/python3.10/site-packages (from matplotlib!=3.6.1,>=3.4->seaborn) (11.2.1)
Requirement already satisfied: fonttools>=4.22.0 in ./myenv3.10/lib/python3.10/site-packages (from matplotlib!=3.6.1,>=3.4->seaborn) (4.58.2)
Requirement already satisfied: python-dateutil>=2.7 in ./myenv3.10/lib/python3.10/site-packages (from matplotlib!=3.6.1,>=3.4->seaborn) (2.9.0.post0)
Requirement already satisfied: pytz>=2020.1 in ./myenv3.10/lib/python3.10/site-packages (from pandas>=1.2->seaborn) (2025.2)
Requirement already satisfied: six>=1.5 in ./myenv3.10/lib/python3.10/site-packages (from python-dateutil>=2.7->matplotlib!=3.6.1,>=3.4->seaborn) (1.17.0)

[notice] A new release of pip available: 22.2.2 -> 25.1.1
[notice] To update, run: pip install --upgrade pip
Note: you may need to restart the kernel to use updated packages.

pip install plotly

Requirement already satisfied: plotly in ./myenv3.10/lib/python3.10/site-packages (6.1.2)
Requirement already satisfied: narwhals>=1.15.1 in ./myenv3.10/lib/python3.10/site-packages (from plotly) (1.42.0)
Requirement already satisfied: packaging in ./myenv3.10/lib/python3.10/site-packages (from plotly) (25.0)

[notice] A new release of pip available: 22.2.2 -> 25.1.1
[notice] To update, run: pip install --upgrade pip
Note: you may need to restart the kernel to use updated packages.

pip install nbformat

Requirement already satisfied: nbformat in ./myenv3.10/lib/python3.10/site-packages (5.10.4)
Requirement already satisfied: traitlets>=5.1 in ./myenv3.10/lib/python3.10/site-packages (from nbformat) (5.14.3)
Requirement already satisfied: jupyter-core!=5.0.*,>=4.12 in ./myenv3.10/lib/python3.10/site-packages (from nbformat) (5.8.1)
Requirement already satisfied: jsonschema>=2.6 in ./myenv3.10/lib/python3.10/site-packages (from nbformat) (4.24.0)
Requirement already satisfied: fastjsonschema>=2.15 in ./myenv3.10/lib/python3.10/site-packages (from nbformat) (2.21.1)
Requirement already satisfied: referencing>=0.28.4 in ./myenv3.10/lib/python3.10/site-packages (from jsonschema>=2.6->nbformat) (0.36.2)
Requirement already satisfied: rpds-py>=0.7.1 in ./myenv3.10/lib/python3.10/site-packages (from jsonschema>=2.6->nbformat) (0.25.1)
Requirement already satisfied: jsonschema-specifications>=2023.03.6 in ./myenv3.10/lib/python3.10/site-packages (from jsonschema>=2.6->nbformat) (2025.4.1)
Requirement already satisfied: attrs>=22.2.0 in ./myenv3.10/lib/python3.10/site-packages (from jsonschema>=2.6->nbformat) (25.3.0)
Requirement already satisfied: platformdirs>=2.5 in ./myenv3.10/lib/python3.10/site-packages (from jupyter-core!=5.0.*,>=4.12->nbformat) (4.3.8)
Requirement already satisfied: typing-extensions>=4.4.0 in ./myenv3.10/lib/python3.10/site-packages (from referencing>=0.28.4->jsonschema>=2.6->nbformat) (4.14.0)

[notice] A new release of pip available: 22.2.2 -> 25.1.1
[notice] To update, run: pip install --upgrade pip
Note: you may need to restart the kernel to use updated packages.

lagged_stats.isnull().sum().to_frame('null_count').T

lagged_stats.describe()

import plotly.graph_objs as go
from plotly.subplots import make_subplots

# Select numeric columns
numeric_cols = lagged_stats.select_dtypes(include='number').columns.tolist()
num_plots = len(cols)

# Define grid size: 11 rows, 5 cols
rows = 10
cols_per_row = 5

# Create subplot figure
fig = make_subplots(rows=rows, cols=cols_per_row,
                    subplot_titles=numeric_cols[:rows*cols_per_row])

for i, col_name in enumerate(numeric_cols[:rows*cols_per_row]):
    r = (i // cols_per_row) + 1
    c = (i % cols_per_row) + 1

    fig.add_trace(
        go.Histogram(x=lagged_stats[col_name], name=col_name, showlegend=False),
        row=r, col=c
    )

fig.update_layout(height=1750, width=1500, title_text="Histograms grid")
fig.show()

import seaborn as sns
import matplotlib.pyplot as plt

# Specify your columns of interest
cols_to_use = lagged_stats.select_dtypes(include='number').columns.tolist()

# Compute correlation matrix
corr = lagged_stats[cols_to_use].corr(method='spearman').round(2) ### ONLY using spearman as target variable is skewed

# Plot heatmap
plt.figure(figsize=(25, 20))
sns.heatmap(corr, annot=True, cmap='coolwarm', vmin=-1, vmax=1)
plt.title('SPEARMAN Correlation Heatmap')
plt.show()

pip install statsmodels

Requirement already satisfied: statsmodels in ./myenv3.10/lib/python3.10/site-packages (0.14.4)
Requirement already satisfied: scipy!=1.9.2,>=1.8 in ./myenv3.10/lib/python3.10/site-packages (from statsmodels) (1.15.3)
Requirement already satisfied: patsy>=0.5.6 in ./myenv3.10/lib/python3.10/site-packages (from statsmodels) (1.0.1)
Requirement already satisfied: packaging>=21.3 in ./myenv3.10/lib/python3.10/site-packages (from statsmodels) (25.0)
Requirement already satisfied: numpy<3,>=1.22.3 in ./myenv3.10/lib/python3.10/site-packages (from statsmodels) (1.26.4)
Requirement already satisfied: pandas!=2.1.0,>=1.4 in ./myenv3.10/lib/python3.10/site-packages (from statsmodels) (1.5.3)
Requirement already satisfied: pytz>=2020.1 in ./myenv3.10/lib/python3.10/site-packages (from pandas!=2.1.0,>=1.4->statsmodels) (2025.2)
Requirement already satisfied: python-dateutil>=2.8.1 in ./myenv3.10/lib/python3.10/site-packages (from pandas!=2.1.0,>=1.4->statsmodels) (2.9.0.post0)
Requirement already satisfied: six>=1.5 in ./myenv3.10/lib/python3.10/site-packages (from python-dateutil>=2.8.1->pandas!=2.1.0,>=1.4->statsmodels) (1.17.0)

[notice] A new release of pip available: 22.2.2 -> 25.1.1
[notice] To update, run: pip install --upgrade pip
Note: you may need to restart the kernel to use updated packages.

# Taking out receiving stats - irrelevant for the most part
qb_eda_df = lagged_stats[lagged_stats['position'] == 'QB'].iloc[:, list(range(0,35)) + list(range(51,56))]
qb_eda_df

qb_eda_df.isnull().sum().to_frame('null_count').T

qb_eda_df.describe()

import plotly.graph_objs as go
from plotly.subplots import make_subplots

# Sample dataframe
df = qb_eda_df  # Replace with your actual df

# Select numeric columns
numeric_cols = qb_eda_df.select_dtypes(include='number').columns.tolist()
num_plots = len(cols)

# Define grid size: 7 rows, 4 cols
rows = 8
cols_per_row = 4

# Create subplot figure
fig = make_subplots(rows=rows, cols=cols_per_row,
                    subplot_titles=numeric_cols[:rows*cols_per_row])

for i, col_name in enumerate(numeric_cols[:rows*cols_per_row]):
    r = (i // cols_per_row) + 1
    c = (i % cols_per_row) + 1

    fig.add_trace(
        go.Histogram(x=df[col_name], name=col_name, showlegend=False),
        row=r, col=c
    )

fig.update_layout(height=1400, width=1200, title_text="Histograms grid")
fig.show()

import seaborn as sns
import matplotlib.pyplot as plt

# Specify your columns of interest
cols_to_use = ['fantasy_points_ppr', 'completions_lagged', 'attempts_lagged', 'passing_yards_lagged', 'passing_air_yards_lagged', 'passing_yards_after_catch_lagged',
                        'passing_first_downs_lagged', 'passing_epa_lagged', 'rushing_epa_lagged', 'fantasy_points_ppr_lagged']

# Compute correlation matrix
corr = qb_eda_df[cols_to_use].corr(method='pearson').round(3)

# Plot heatmap
plt.figure(figsize=(10, 8))
sns.heatmap(corr, annot=True, cmap='coolwarm', vmin=-1, vmax=1)
plt.title('PEARSON Correlation Heatmap')
plt.show()

import seaborn as sns
import matplotlib.pyplot as plt

# Specify your columns of interest
cols_to_use = qb_eda_df.select_dtypes(include='number').drop(columns=['completions_lagged', 'attempts_lagged', 
                        'passing_yards_lagged', 'passing_air_yards_lagged', 'passing_yards_after_catch_lagged',
                        'passing_first_downs_lagged', 'passing_epa_lagged', 'rushing_epa_lagged', 'fantasy_points_ppr_lagged', 'used_opponent_avg', 'used_game_number_avg',
                        'DNP_prior_season_game', 'never_seen']).columns.tolist()

# Compute correlation matrix
corr = qb_eda_df[cols_to_use].corr(method='spearman').round(3) ### Looking at spearman as well because some variables are skewed

# Plot heatmap
plt.figure(figsize=(15, 12))
sns.heatmap(corr, annot=True, cmap='coolwarm', vmin=-1, vmax=1)
plt.title('SPEARMAN Correlation Heatmap')
plt.show()

# Only keeping receiving stats - rushing/passing irrelevant for the most part
wr_eda_df = lagged_stats[lagged_stats['position'] == 'WR'].iloc[:, list(range(0,11)) + list(range(35, 56))]
wr_eda_df

wr_eda_df.isnull().sum().to_frame('null_count').T

wr_eda_df.describe()

import plotly.graph_objs as go
from plotly.subplots import make_subplots

# Sample dataframe
df = wr_eda_df  # Replace with your actual df

# Select numeric columns
numeric_cols = wr_eda_df.select_dtypes(include='number').columns.tolist()
num_plots = len(cols)

# Define grid size: 7 rows, 4 cols
rows = 6
cols_per_row = 4

# Create subplot figure
fig = make_subplots(rows=rows, cols=cols_per_row,
                    subplot_titles=numeric_cols[:rows*cols_per_row])

for i, col_name in enumerate(numeric_cols[:rows*cols_per_row]):
    r = (i // cols_per_row) + 1
    c = (i % cols_per_row) + 1

    fig.add_trace(
        go.Histogram(x=df[col_name], name=col_name, showlegend=False),
        row=r, col=c
    )

fig.update_layout(height=1400, width=1200, title_text="Histograms grid")
fig.show()

import seaborn as sns
import matplotlib.pyplot as plt

# Specify your columns of interest
cols_to_use = wr_eda_df.select_dtypes(include='number').columns.tolist()

# Compute correlation matrix
corr = wr_eda_df[cols_to_use].corr(method='spearman').round(3) ### ONLY looking at spearman because target variable is skewed

# Plot heatmap
plt.figure(figsize=(20, 16))
sns.heatmap(corr, annot=True, cmap='coolwarm', vmin=-1, vmax=1)
plt.title("SPEARMAN Correlation Heatmap")
plt.show()

# Only keeping rushing/receiving stats - passing irrelevant for the most part
rb_eda_df = lagged_stats[lagged_stats['position'].isin(['RB', 'FB'])].iloc[:, list(range(0,11)) + list(range(27, 56))]
rb_eda_df

rb_eda_df.isnull().sum().to_frame('null_count').T

rb_eda_df.describe()

import plotly.graph_objs as go
from plotly.subplots import make_subplots

# Sample dataframe
df = rb_eda_df  # Replace with your actual df

# Select numeric columns
numeric_cols = rb_eda_df.select_dtypes(include='number').columns.tolist()
num_plots = len(cols)

# Define grid size: 7 rows, 4 cols
rows = 8
cols_per_row = 4

# Create subplot figure
fig = make_subplots(rows=rows, cols=cols_per_row,
                    subplot_titles=numeric_cols[:rows*cols_per_row])

for i, col_name in enumerate(numeric_cols[:rows*cols_per_row]):
    r = (i // cols_per_row) + 1
    c = (i % cols_per_row) + 1

    fig.add_trace(
        go.Histogram(x=df[col_name], name=col_name, showlegend=False),
        row=r, col=c
    )

fig.update_layout(height=1400, width=1200, title_text="Histograms grid")
fig.show()

import seaborn as sns
import matplotlib.pyplot as plt

# Specify your columns of interest
cols_to_use = rb_eda_df.select_dtypes(include='number').columns.tolist()

# Compute correlation matrix
corr = rb_eda_df[cols_to_use].corr(method='spearman').round(3) ### ONLY looking at spearman because target variable is skewed

# Plot heatmap
plt.figure(figsize=(20, 16))
sns.heatmap(corr, annot=True, cmap='coolwarm', vmin=-1, vmax=1)
plt.title("SPEARMAN Correlation Heatmap")
plt.show()

# Only keeping receiving stats - rushing/passing irrelevant for the most part
te_eda_df = lagged_stats[lagged_stats['position'] == 'TE'].iloc[:, list(range(0,11)) + list(range(35, 56))]
te_eda_df

te_eda_df.isnull().sum().to_frame('null_count').T

te_eda_df.describe()

import plotly.graph_objs as go
from plotly.subplots import make_subplots

# Sample dataframe
df = te_eda_df  # Replace with your actual df

# Select numeric columns
numeric_cols = te_eda_df.select_dtypes(include='number').columns.tolist()
num_plots = len(cols)

# Define grid size: 7 rows, 4 cols
rows = 6
cols_per_row = 4

# Create subplot figure
fig = make_subplots(rows=rows, cols=cols_per_row,
                    subplot_titles=numeric_cols[:rows*cols_per_row])

for i, col_name in enumerate(numeric_cols[:rows*cols_per_row]):
    r = (i // cols_per_row) + 1
    c = (i % cols_per_row) + 1

    fig.add_trace(
        go.Histogram(x=df[col_name], name=col_name, showlegend=False),
        row=r, col=c
    )

fig.update_layout(height=1400, width=1200, title_text="Histograms grid")
fig.show()

import seaborn as sns
import matplotlib.pyplot as plt

# Specify your columns of interest
cols_to_use = te_eda_df.select_dtypes(include='number').columns.tolist()

# Compute correlation matrix
corr = te_eda_df[cols_to_use].corr(method='spearman').round(3) ### ONLY looking at spearman because target variable is skewed

# Plot heatmap
plt.figure(figsize=(20, 16))
sns.heatmap(corr, annot=True, cmap='coolwarm', vmin=-1, vmax=1)
plt.title("SPEARMAN Correlation Heatmap")
plt.show()

pip install catboost

Requirement already satisfied: catboost in ./myenv3.10/lib/python3.10/site-packages (1.2.8)
Requirement already satisfied: matplotlib in ./myenv3.10/lib/python3.10/site-packages (from catboost) (3.10.3)
Requirement already satisfied: scipy in ./myenv3.10/lib/python3.10/site-packages (from catboost) (1.15.3)
Requirement already satisfied: six in ./myenv3.10/lib/python3.10/site-packages (from catboost) (1.17.0)
Requirement already satisfied: pandas>=0.24 in ./myenv3.10/lib/python3.10/site-packages (from catboost) (1.5.3)
Requirement already satisfied: numpy<3.0,>=1.16.0 in ./myenv3.10/lib/python3.10/site-packages (from catboost) (1.26.4)
Requirement already satisfied: plotly in ./myenv3.10/lib/python3.10/site-packages (from catboost) (6.1.2)
Requirement already satisfied: graphviz in ./myenv3.10/lib/python3.10/site-packages (from catboost) (0.20.3)
Requirement already satisfied: pytz>=2020.1 in ./myenv3.10/lib/python3.10/site-packages (from pandas>=0.24->catboost) (2025.2)
Requirement already satisfied: python-dateutil>=2.8.1 in ./myenv3.10/lib/python3.10/site-packages (from pandas>=0.24->catboost) (2.9.0.post0)
Requirement already satisfied: pillow>=8 in ./myenv3.10/lib/python3.10/site-packages (from matplotlib->catboost) (11.2.1)
Requirement already satisfied: contourpy>=1.0.1 in ./myenv3.10/lib/python3.10/site-packages (from matplotlib->catboost) (1.3.2)
Requirement already satisfied: fonttools>=4.22.0 in ./myenv3.10/lib/python3.10/site-packages (from matplotlib->catboost) (4.58.2)
Requirement already satisfied: kiwisolver>=1.3.1 in ./myenv3.10/lib/python3.10/site-packages (from matplotlib->catboost) (1.4.8)
Requirement already satisfied: cycler>=0.10 in ./myenv3.10/lib/python3.10/site-packages (from matplotlib->catboost) (0.12.1)
Requirement already satisfied: packaging>=20.0 in ./myenv3.10/lib/python3.10/site-packages (from matplotlib->catboost) (25.0)
Requirement already satisfied: pyparsing>=2.3.1 in ./myenv3.10/lib/python3.10/site-packages (from matplotlib->catboost) (3.2.3)
Requirement already satisfied: narwhals>=1.15.1 in ./myenv3.10/lib/python3.10/site-packages (from plotly->catboost) (1.42.0)

[notice] A new release of pip available: 22.2.2 -> 25.1.1
[notice] To update, run: pip install --upgrade pip
Note: you may need to restart the kernel to use updated packages.

pip install scikit-learn

Requirement already satisfied: scikit-learn in ./myenv3.10/lib/python3.10/site-packages (1.7.0)
Requirement already satisfied: numpy>=1.22.0 in ./myenv3.10/lib/python3.10/site-packages (from scikit-learn) (1.26.4)
Requirement already satisfied: threadpoolctl>=3.1.0 in ./myenv3.10/lib/python3.10/site-packages (from scikit-learn) (3.6.0)
Requirement already satisfied: scipy>=1.8.0 in ./myenv3.10/lib/python3.10/site-packages (from scikit-learn) (1.15.3)
Requirement already satisfied: joblib>=1.2.0 in ./myenv3.10/lib/python3.10/site-packages (from scikit-learn) (1.5.1)

[notice] A new release of pip available: 22.2.2 -> 25.1.1
[notice] To update, run: pip install --upgrade pip
Note: you may need to restart the kernel to use updated packages.

from sklearn.preprocessing import StandardScaler
from sklearn.model_selection import train_test_split, RandomizedSearchCV
from sklearn.metrics import mean_absolute_error, r2_score, mean_squared_error, mean_absolute_percentage_error
from catboost import CatBoostRegressor, cv, Pool
from itertools import product
from tqdm import tqdm

# Training split
qb_train = qb_eda_df.sample(n=round(len(qb_eda_df)*.7), random_state=12)
qb_train_X = qb_train.loc[:, qb_train.columns!='fantasy_points_ppr']
qb_train_y = qb_train['fantasy_points_ppr']
qb_train_indices = qb_train.index

qb_nontrain = qb_eda_df.drop(qb_train.index)

# Test split
qb_test = qb_nontrain.sample(n=round(len(qb_nontrain)*.5), random_state=12)
qb_test_X = qb_test.loc[:, qb_test.columns!='fantasy_points_ppr']
qb_test_y = qb_test['fantasy_points_ppr']
qb_test_X_indices = qb_test_X.index

# Validation split
qb_val = qb_nontrain.drop(qb_test.index)
qb_val_X = qb_val.loc[:, qb_val.columns!='fantasy_points_ppr']
qb_val_y = qb_val['fantasy_points_ppr']
qb_val

# Parameter grid
iterations_ls = [500, 1000]
learning_rates_ls = [0.01, 0.03, 0.1]
depth_list = [4,6,8]

qb_cv_results = []

qb_cvpool = Pool(data=qb_train_X, label=qb_train_y, cat_features=['player_id', 'player_name', 'position', 'team', 'opponent_team', 'day_slate', 'location'])

# Grid search
for i, l, d in tqdm(product(iterations_ls, learning_rates_ls, depth_list),
                    total = len(iterations_ls)*len(learning_rates_ls)*len(depth_list)):
        qb_params = {'loss_function':'MAE',
                     'iterations': i,          
                     'learning_rate': l,       
                     'depth': d,
                     'early_stopping_rounds': 50,
                     'verbose': False, }                  
        qb_cvscores = cv(pool=qb_cvpool, params = qb_params, fold_count = 5, shuffle=True, logging_level='Silent', partition_random_seed=11)
        best_qb_cvscore = qb_cvscores['test-MAE-mean'].min()
        qb_cv_results.append({'iterations': i,
                              'learning_rate': l,
                              'depth': d,
                              'mae': best_qb_cvscore})

qb_cv_results_df = pd.DataFrame(qb_cv_results)
best_params_qb = qb_cv_results_df.sort_values(by='mae').iloc[0]
best_params_qb

  0%|          | 0/18 [00:00<?, ?it/s]

100%|██████████| 18/18 [05:19<00:00, 17.77s/it]

iterations       1000.000000
learning_rate       0.030000
depth               6.000000
mae                 6.108728
Name: 13, dtype: float64

# Training Model on Full Training Set
final_qb_params = {
    'loss_function':'MAE',
    'iterations': int(best_params_qb['iterations']),
    'learning_rate': float(best_params_qb['learning_rate']),
    'depth': int(best_params_qb['depth']),
    'early_stopping_rounds': 50,
    'verbose': False
}

qb_catboost = CatBoostRegressor(**final_qb_params)
qb_catboost.fit(qb_train_X, qb_train_y, cat_features=['player_id', 'player_name', 'position', 'team', 'opponent_team', 'day_slate', 'location'], 
                eval_set=(qb_val_X, qb_val_y))

<catboost.core.CatBoostRegressor at 0x12d8ba320>

qb_val_preds = qb_catboost.predict(qb_val_X)
print(f'Validation MAE: {mean_absolute_error(qb_val_y, qb_val_preds)}')
print(f'Validation R2: {r2_score(qb_val_y, qb_val_preds)}')

Validation MAE: 5.955763733268118
Validation R2: 0.2575433842264834

qb_test_preds = qb_catboost.predict(qb_test_X)
print(f'Test MAE: {mean_absolute_error(qb_test_y, qb_test_preds)}')
print(f'Test R2: {r2_score(qb_test_y, qb_test_preds)}')

# Predictions with indices
qb_test_preds_wind = np.column_stack((qb_test_X_indices, qb_test_preds))
qb_sorted_preds = qb_test_preds_wind[qb_test_preds_wind[:,0].argsort()]
qb_sorted_preds_df = pd.DataFrame(qb_sorted_preds, columns = ['index', 'prediction'])
qb_sorted_preds_df.set_index('index', inplace=True)
qb_sorted_preds_v_test = pd.concat([qb_test_X, qb_sorted_preds_df], axis = 1)
qb_test_preds_exp = pd.concat([qb_sorted_preds_v_test, qb_test_y], axis=1)
qb_test_preds_exp

Test MAE: 5.939019054381559
Test R2: 0.31621628792198975

pip install shap

Requirement already satisfied: shap in ./myenv3.10/lib/python3.10/site-packages (0.47.2)
Requirement already satisfied: scikit-learn in ./myenv3.10/lib/python3.10/site-packages (from shap) (1.7.0)
Requirement already satisfied: numpy in ./myenv3.10/lib/python3.10/site-packages (from shap) (1.26.4)
Requirement already satisfied: cloudpickle in ./myenv3.10/lib/python3.10/site-packages (from shap) (3.1.1)
Requirement already satisfied: packaging>20.9 in ./myenv3.10/lib/python3.10/site-packages (from shap) (25.0)
Requirement already satisfied: slicer==0.0.8 in ./myenv3.10/lib/python3.10/site-packages (from shap) (0.0.8)
Requirement already satisfied: typing-extensions in ./myenv3.10/lib/python3.10/site-packages (from shap) (4.14.0)
Requirement already satisfied: scipy in ./myenv3.10/lib/python3.10/site-packages (from shap) (1.15.3)
Requirement already satisfied: tqdm>=4.27.0 in ./myenv3.10/lib/python3.10/site-packages (from shap) (4.67.1)
Requirement already satisfied: numba>=0.54 in ./myenv3.10/lib/python3.10/site-packages (from shap) (0.61.2)
Requirement already satisfied: pandas in ./myenv3.10/lib/python3.10/site-packages (from shap) (1.5.3)
Requirement already satisfied: llvmlite<0.45,>=0.44.0dev0 in ./myenv3.10/lib/python3.10/site-packages (from numba>=0.54->shap) (0.44.0)
Requirement already satisfied: python-dateutil>=2.8.1 in ./myenv3.10/lib/python3.10/site-packages (from pandas->shap) (2.9.0.post0)
Requirement already satisfied: pytz>=2020.1 in ./myenv3.10/lib/python3.10/site-packages (from pandas->shap) (2025.2)
Requirement already satisfied: joblib>=1.2.0 in ./myenv3.10/lib/python3.10/site-packages (from scikit-learn->shap) (1.5.1)
Requirement already satisfied: threadpoolctl>=3.1.0 in ./myenv3.10/lib/python3.10/site-packages (from scikit-learn->shap) (3.6.0)
Requirement already satisfied: six>=1.5 in ./myenv3.10/lib/python3.10/site-packages (from python-dateutil>=2.8.1->pandas->shap) (1.17.0)

[notice] A new release of pip available: 22.2.2 -> 25.1.1
[notice] To update, run: pip install --upgrade pip
Note: you may need to restart the kernel to use updated packages.

import shap
explainer_qb = shap.TreeExplainer(qb_catboost, feature_perturbation='tree_path_dependent')
shap_values_qb = explainer_qb(qb_test_X)
shap_values_qb

/Users/kevingarand/Desktop/VS code python/myenv3.10/lib/python3.10/site-packages/tqdm/auto.py:21: TqdmWarning:

IProgress not found. Please update jupyter and ipywidgets. See https://ipywidgets.readthedocs.io/en/stable/user_install.html

.values =
array([[ 0.12718426, -0.02909622,  0.01079839, ..., -0.00625475,
        -0.05422574,  0.        ],
       [-3.61396261, -2.7365059 ,  0.04298625, ...,  0.04234179,
         0.02432914,  0.        ],
       [ 0.74639142,  0.06668904,  0.06701884, ..., -0.00861325,
        -0.02255906,  0.        ],
       ...,
       [ 0.98696398,  0.56557576, -0.04998133, ...,  0.00643863,
         0.01213931,  0.        ],
       [-0.27180811,  0.60021587, -0.065277  , ..., -0.00959056,
        -0.03073108,  0.        ],
       [ 1.84438444,  1.71662713,  0.05957254, ..., -0.01529152,
         0.03215778,  0.        ]])

.base_values =
array([13.35321344, 13.35321344, 13.35321344, ..., 13.35321344,
       13.35321344, 13.35321344])

.data =
array([['00-0034869', 'Sam Darnold', '13.0', ..., 1, 1, 0],
       ['00-0024296', 'Charlie Whitehurst', '13.0', ..., 0, 0, 0],
       ['00-0024226', 'Jay Cutler', '8.0', ..., 1, 1, 0],
       ...,
       ['00-0035710', 'Daniel Jones', '3.0', ..., 0, 0, 0],
       ['00-0037077', 'Sam Howell', '7.0', ..., 1, 1, 1],
       ['00-0019596', 'Tom Brady', '9.0', ..., 0, 0, 0]], dtype=object)

shap.plots.beeswarm(shap_values_qb, max_display = 25)

# Exploring causes for errors
qb_test_preds_exp['AE'] = abs(qb_test_preds_exp['prediction'] - qb_test_preds_exp['fantasy_points_ppr'])

print(f"% of instances where absolute error > 6 and QB has never been seen: {100 * len(qb_test_preds_exp[(qb_test_preds_exp['AE'] > 6) & (qb_test_preds_exp['never_seen'] == 1)]) / len(qb_test_preds_exp[qb_test_preds_exp['AE'] > 6]):.2f}%")
print(f"% of instances where absolute error > 6 and QB did not play in prior season game number (seen before though): {100 * len(qb_test_preds_exp[(qb_test_preds_exp['AE'] > 6) & (qb_test_preds_exp['DNP_prior_season_game'] == 1) & (qb_test_preds_exp['never_seen'] != 1)]) / len(qb_test_preds_exp[qb_test_preds_exp['AE'] > 6]):.2f}%")
print(f"Top 5 teams of QBs with largest MAE: {qb_test_preds_exp[['team', 'prediction', 'AE']].groupby(['team']).mean().sort_values('AE', ascending=False).reset_index()['team'].head(5).tolist()}")
print(f"Top 5 game locations of QBs with largest MAE: {qb_test_preds_exp[['location', 'prediction', 'AE']].groupby(['location']).mean().sort_values('AE', ascending=False).reset_index()['location'].head(5).tolist()}")
print(f"Top 5 game numbers of QBs with largest MAE: {qb_test_preds_exp[['game_number', 'prediction', 'AE']].groupby(['game_number']).mean().sort_values('AE', ascending=False).reset_index()['game_number'].head(5).tolist()}")
print(f"Top 5 seasons of QBs with largest MAE: {qb_test_preds_exp[['season', 'prediction', 'AE']].groupby(['season']).mean().sort_values('AE', ascending=False).reset_index()['season'].head(5).tolist()}")
print(f"Top 10 QBs with largest MAE: {qb_test_preds_exp[['player_name', 'prediction', 'AE']].groupby(['player_name']).mean().sort_values('AE', ascending=False).reset_index()['player_name'].head(10).tolist()}")

% of instances where absolute error > 6 and QB has never been seen: 22.52%
% of instances where absolute error > 6 and QB did not play in prior season game number (seen before though): 18.51%
Top 5 teams of QBs with largest MAE: ['GB', 'WAS', 'NYG', 'PHI', 'BUF']
Top 5 game locations of QBs with largest MAE: ['DAL', 'GB', 'PHI', 'NYG', 'NYJ']
Top 5 game numbers of QBs with largest MAE: ['8.0', '16.0', '10.0', '4.0', '6.0']
Top 5 seasons of QBs with largest MAE: ['2019', '2015', '2020', '2021', '2023']
Top 10 QBs with largest MAE: ['Malik Willis', 'Brett Favre', 'Jon Kitna', 'Bailey Zappe', 'Stephen McGee', 'Jake Browning', 'Sam Howell', 'Matt Moore', 'Drew Lock', 'Drake Maye']

qb_season = lagged_stats[lagged_stats['position'] == 'QB'].groupby(['player_name', 'season']).agg(
    {'fantasy_points_ppr':'sum',
     'never_seen':'mean',
     'DNP_prior_season_game':'mean'}
).reset_index()
qb_season.describe()
## Median of target variable = ~70 points, so do QBs with 70+ points in a fantasy season only

qb_filter_df = qb_season[(qb_season['fantasy_points_ppr'] >= 70) & (qb_season['never_seen'] != 1) & (qb_season['DNP_prior_season_game'] != 1)]
qbs_filtered = qb_eda_df.merge(qb_filter_df[['player_name', 'season']], how='inner', on=['player_name', 'season'])
qbs_filtered

# Training split
qbf_train = qbs_filtered.sample(n=round(len(qbs_filtered)*.7), random_state=10)
qbf_train_X = qbf_train.loc[:, qbf_train.columns!='fantasy_points_ppr']
qbf_train_y = qbf_train['fantasy_points_ppr']
qbf_train_indices = qbf_train.index

qbf_nontrain = qbs_filtered.drop(qbf_train.index)

# Test split
qbf_test = qbf_nontrain.sample(n=round(len(qbf_nontrain)*.5), random_state=10)
qbf_test_X = qbf_test.loc[:, qbf_test.columns!='fantasy_points_ppr']
qbf_test_y = qbf_test['fantasy_points_ppr']
qbf_test_X_indices = qbf_test_X.index

# Validation split
qbf_val = qbf_nontrain.drop(qbf_test.index)
qbf_val_X = qbf_val.loc[:, qbf_val.columns!='fantasy_points_ppr']
qbf_val_y = qbf_val['fantasy_points_ppr']
qbf_val

# Parameter grid
iterations_ls = [500, 1000]
learning_rates_ls = [0.01, 0.03, 0.1]
depth_list = [4,6,8]

qbf_cv_results = []

qbf_cvpool = Pool(data=qbf_train_X, label=qbf_train_y, cat_features=['player_id', 'player_name', 'position', 'team', 'opponent_team', 'day_slate', 'location'])

# Grid search
for i, l, d in tqdm(product(iterations_ls, learning_rates_ls, depth_list),
                    total = len(iterations_ls)*len(learning_rates_ls)*len(depth_list)):
        qb_params = {'loss_function':'MAE',
                     'iterations': i,          
                     'learning_rate': l,       
                     'depth': d,
                     'early_stopping_rounds': 50,
                     'verbose': False, }                  
        qbf_cvscores = cv(pool=qbf_cvpool, params = qb_params, fold_count = 5, shuffle=True, logging_level='Silent', partition_random_seed=11)
        best_qbf_cvscore = qbf_cvscores['test-MAE-mean'].min()
        qbf_cv_results.append({'iterations': i,
                              'learning_rate': l,
                              'depth': d,
                              'mae': best_qbf_cvscore})

qbf_cv_results_df = pd.DataFrame(qbf_cv_results)
best_params_qbf = qbf_cv_results_df.sort_values(by='mae').iloc[0]
best_params_qbf

  0%|          | 0/18 [00:00<?, ?it/s]

100%|██████████| 18/18 [03:42<00:00, 12.36s/it]

iterations       500.000000
learning_rate      0.030000
depth              4.000000
mae                6.069722
Name: 3, dtype: float64

# Training Model on Full Training Set
final_qbf_params = {
    'loss_function':'MAE',
    'iterations': int(best_params_qbf['iterations']),
    'learning_rate': float(best_params_qbf['learning_rate']),
    'depth': int(best_params_qbf['depth']),
    'early_stopping_rounds': 50,
    'verbose': False
}

qbf_catboost = CatBoostRegressor(**final_qbf_params)
qbf_catboost.fit(qbf_train_X, qbf_train_y, cat_features=['player_id', 'player_name', 'position', 'team', 'opponent_team', 'day_slate', 'location'], 
                eval_set=(qbf_val_X, qbf_val_y))

<catboost.core.CatBoostRegressor at 0x12fdc2ad0>

qbf_val_preds = qbf_catboost.predict(qbf_val_X)
print(f'Validation MAE: {mean_absolute_error(qbf_val_y, qbf_val_preds)}')
print(f'Validation R2: {r2_score(qbf_val_y, qbf_val_preds)}')

Validation MAE: 5.997673176453284
Validation R2: 0.11752783643652409

qbf_test_preds = qbf_catboost.predict(qbf_test_X)
print(f'Test MAE: {mean_absolute_error(qbf_test_y, qbf_test_preds)}')
print(f'Test R2: {r2_score(qbf_test_y, qbf_test_preds)}')

# Predictions with indices
qbf_test_preds_wind = np.column_stack((qbf_test_X_indices, qbf_test_preds))
qbf_sorted_preds = qbf_test_preds_wind[qbf_test_preds_wind[:,0].argsort()]
qbf_sorted_preds_df = pd.DataFrame(qbf_sorted_preds, columns = ['index', 'prediction'])
qbf_sorted_preds_df.set_index('index', inplace=True)
qbf_sorted_preds_v_test = pd.concat([qbf_test_X, qbf_sorted_preds_df], axis = 1)
qbf_test_preds_exp = pd.concat([qbf_sorted_preds_v_test, qbf_test_y], axis=1)
qbf_test_preds_exp

Test MAE: 5.952722080726591
Test R2: 0.09677953695651109

# Training split
wr_train = wr_eda_df.sample(n=round(len(wr_eda_df)*.7), random_state=18)
wr_train_X = wr_train.loc[:, wr_train.columns!='fantasy_points_ppr']
wr_train_y = wr_train['fantasy_points_ppr']
wr_train_indices = wr_train.index

wr_nontrain = wr_eda_df.drop(wr_train.index)

# Test split
wr_test = wr_nontrain.sample(n=round(len(wr_nontrain)*.5), random_state=18)
wr_test_X = wr_test.loc[:, wr_test.columns!='fantasy_points_ppr']
wr_test_y = wr_test['fantasy_points_ppr']
wr_test_X_indices = wr_test_X.index

# Validation split
wr_val = wr_nontrain.drop(wr_test.index)
wr_val_X = wr_val.loc[:, wr_val.columns!='fantasy_points_ppr']
wr_val_y = wr_val['fantasy_points_ppr']
wr_val

# Parameter grid
iterations_ls = [500, 1000]
learning_rates_ls = [0.01, 0.03, 0.1]
depth_list = [4,6,8]

wr_cv_results = []

wr_cvpool = Pool(data=wr_train_X, label=wr_train_y, cat_features=['player_id', 'player_name', 'position', 'team', 'opponent_team', 'day_slate', 'location'])

# Grid search
for i, l, d in tqdm(product(iterations_ls, learning_rates_ls, depth_list),
                    total = len(iterations_ls)*len(learning_rates_ls)*len(depth_list)):
        wr_params = {'loss_function':'MAE',
                     'iterations': i,          
                     'learning_rate': l,       
                     'depth': d,
                     'early_stopping_rounds': 50,
                     'verbose': False, }                  
        wr_cvscores = cv(pool=wr_cvpool, params = wr_params, fold_count = 5, shuffle=True, logging_level='Silent', partition_random_seed=18)
        best_wr_cvscore = wr_cvscores['test-MAE-mean'].min()
        wr_cv_results.append({'iterations': i,
                              'learning_rate': l,
                              'depth': d,
                              'mae': best_wr_cvscore})

wr_cv_results_df = pd.DataFrame(wr_cv_results)
best_params_wr = wr_cv_results_df.sort_values(by='mae').iloc[0]
best_params_wr

100%|██████████| 18/18 [08:24<00:00, 28.04s/it]

iterations       1000.00000
learning_rate       0.01000
depth               8.00000
mae                 4.98711
Name: 11, dtype: float64

# Training Model on Full Training Set
final_wr_params = {
    'loss_function':'MAE',
    'iterations': int(best_params_wr['iterations']),
    'learning_rate': float(best_params_wr['learning_rate']),
    'depth': int(best_params_wr['depth']),
    'early_stopping_rounds': 50,
    'verbose': False
}

wr_catboost = CatBoostRegressor(**final_wr_params)
wr_catboost.fit(wr_train_X, wr_train_y, cat_features=['player_id', 'player_name', 'position', 'team', 'opponent_team', 'day_slate', 'location'], 
                eval_set=(wr_val_X, wr_val_y))

<catboost.core.CatBoostRegressor at 0x12fd6ea10>

wr_val_preds = wr_catboost.predict(wr_val_X)
print(f'Validation MAE: {mean_absolute_error(wr_val_y, wr_val_preds)}')
print(f'Validation R2: {r2_score(wr_val_y, wr_val_preds)}')

Validation MAE: 5.055909806491361
Validation R2: 0.22335346606500162

wr_test_preds = wr_catboost.predict(wr_test_X)
print(f'Test MAE: {mean_absolute_error(wr_test_y, wr_test_preds)}')
print(f'Test R2: {r2_score(wr_test_y, wr_test_preds)}')

# Predictions with indices
wr_test_preds_wind = np.column_stack((wr_test_X_indices, wr_test_preds))
wr_sorted_preds = wr_test_preds_wind[wr_test_preds_wind[:,0].argsort()]
wr_sorted_preds_df = pd.DataFrame(wr_sorted_preds, columns = ['index', 'prediction'])
wr_sorted_preds_df.set_index('index', inplace=True)
wr_sorted_preds_v_test = pd.concat([wr_test_X, wr_sorted_preds_df], axis = 1)
wr_test_preds_exp = pd.concat([wr_sorted_preds_v_test, wr_test_y], axis=1)
wr_test_preds_exp

Test MAE: 4.974293749386388
Test R2: 0.23185263042535875

import shap
explainer_wr = shap.TreeExplainer(wr_catboost, feature_perturbation='tree_path_dependent')
shap_values_wr = explainer_wr(wr_test_X)
shap_values_wr

.values =
array([[-0.69830458, -0.83741373, -0.05679496, ..., -0.00612945,
        -0.0513467 ,  0.        ],
       [ 1.01245883,  1.18579801,  0.05486811, ..., -0.00345628,
        -0.11083167,  0.        ],
       [ 0.00663768, -0.09179858,  0.07614193, ...,  0.00935648,
         0.03508649,  0.        ],
       ...,
       [ 2.79426841,  2.52643826,  0.0178823 , ...,  0.00531601,
        -0.14469944,  0.        ],
       [ 2.76191841,  2.43458291, -0.0574296 , ..., -0.01076694,
         0.02864976,  0.        ],
       [-0.29674974, -0.23489613, -0.01230635, ..., -0.00455926,
         0.02181111,  0.        ]])

.base_values =
array([7.32176643, 7.32176643, 7.32176643, ..., 7.32176643, 7.32176643,
       7.32176643])

.data =
array([['00-0033307', 'Kendrick Bourne', '14.0', ..., 1, 1, 1],
       ['00-0027681', 'Brandon LaFell', '3.0', ..., 1, 1, 0],
       ['00-0029000', 'Cole Beasley', '1.0', ..., 0, 0, 0],
       ...,
       ['00-0030279', 'Keenan Allen', '5.0', ..., 1, 1, 0],
       ['00-0036358', 'CeeDee Lamb', '13.0', ..., 1, 0, 0],
       ['00-0026937', 'Brandon Gibson', '9.0', ..., 1, 0, 0]],
      dtype=object)

shap.plots.beeswarm(shap_values_wr, max_display = 25)

# Training split
rb_train = rb_eda_df.sample(n=round(len(rb_eda_df)*.7), random_state=30)
rb_train_X = rb_train.loc[:, rb_train.columns!='fantasy_points_ppr']
rb_train_y = rb_train['fantasy_points_ppr']
rb_train_indices = rb_train.index

rb_nontrain = rb_eda_df.drop(rb_train.index)

# Test split
rb_test = rb_nontrain.sample(n=round(len(rb_nontrain)*.5), random_state=30)
rb_test_X = rb_test.loc[:, rb_test.columns!='fantasy_points_ppr']
rb_test_y = rb_test['fantasy_points_ppr']
rb_test_X_indices = rb_test_X.index

# Validation split
rb_val = rb_nontrain.drop(rb_test.index)
rb_val_X = rb_val.loc[:, rb_val.columns!='fantasy_points_ppr']
rb_val_y = rb_val['fantasy_points_ppr']
rb_val

# Parameter grid
iterations_ls = [500, 1000]
learning_rates_ls = [0.01, 0.03, 0.1]
depth_list = [4,6,8]

rb_cv_results = []

rb_cvpool = Pool(data=rb_train_X, label=rb_train_y, cat_features=['player_id', 'player_name', 'position', 'team', 'opponent_team', 'day_slate', 'location'])

# Grid search
for i, l, d in tqdm(product(iterations_ls, learning_rates_ls, depth_list),
                    total = len(iterations_ls)*len(learning_rates_ls)*len(depth_list)):
        rb_params = {'loss_function':'MAE',
                     'iterations': i,          
                     'learning_rate': l,       
                     'depth': d,
                     'early_stopping_rounds': 50,
                     'verbose': False, }                  
        rb_cvscores = cv(pool=rb_cvpool, params = rb_params, fold_count = 5, shuffle=True, logging_level='Silent', partition_random_seed=30)
        best_rb_cvscore = rb_cvscores['test-MAE-mean'].min()
        rb_cv_results.append({'iterations': i,
                              'learning_rate': l,
                              'depth': d,
                              'mae': best_rb_cvscore})

rb_cv_results_df = pd.DataFrame(rb_cv_results)
best_params_rb = rb_cv_results_df.sort_values(by='mae').iloc[0]
best_params_rb

100%|██████████| 18/18 [08:47<00:00, 29.30s/it]

iterations       1000.000000
learning_rate       0.030000
depth               6.000000
mae                 4.677777
Name: 13, dtype: float64

# Training Model on Full Training Set
final_rb_params = {
    'loss_function':'MAE',
    'iterations': int(best_params_rb['iterations']),
    'learning_rate': float(best_params_rb['learning_rate']),
    'depth': int(best_params_rb['depth']),
    'early_stopping_rounds': 50,
    'verbose': False
}

rb_catboost = CatBoostRegressor(**final_rb_params)
rb_catboost.fit(rb_train_X, rb_train_y, cat_features=['player_id', 'player_name', 'position', 'team', 'opponent_team', 'day_slate', 'location'], 
                eval_set=(rb_val_X, rb_val_y))

<catboost.core.CatBoostRegressor at 0x12ffd3310>

rb_val_preds = rb_catboost.predict(rb_val_X)
print(f'Validation MAE: {mean_absolute_error(rb_val_y, rb_val_preds)}')
print(f'Validation R2: {r2_score(rb_val_y, rb_val_preds)}')

Validation MAE: 4.654278213949994
Validation R2: 0.28995476588807134

rb_test_preds = rb_catboost.predict(rb_test_X)
print(f'Test MAE: {mean_absolute_error(rb_test_y, rb_test_preds)}')
print(f'Test R2: {r2_score(rb_test_y, rb_test_preds)}')

# Predictions with indices
rb_test_preds_wind = np.column_stack((rb_test_X_indices, rb_test_preds))
rb_sorted_preds = rb_test_preds_wind[rb_test_preds_wind[:,0].argsort()]
rb_sorted_preds_df = pd.DataFrame(rb_sorted_preds, columns = ['index', 'prediction'])
rb_sorted_preds_df.set_index('index', inplace=True)
rb_sorted_preds_v_test = pd.concat([rb_test_X, rb_sorted_preds_df], axis = 1)
rb_test_preds_exp = pd.concat([rb_sorted_preds_v_test, rb_test_y], axis=1)
rb_test_preds_exp

Test MAE: 4.630191167189465
Test R2: 0.2870807451387195

explainer_rb = shap.TreeExplainer(rb_catboost, feature_perturbation='tree_path_dependent')
shap_values_rb = explainer_rb(rb_test_X)
shap_values_rb

.values =
array([[-8.90144498e-01, -1.85623913e+00, -1.02817668e-01, ...,
         7.36117638e-04,  5.06839548e-03,  0.00000000e+00],
       [-9.71368087e-01, -2.43072789e+00,  2.58798842e-04, ...,
         2.49013869e-03, -2.36623576e-03,  0.00000000e+00],
       [ 1.08222850e-01,  1.51540663e+00, -2.43688169e-02, ...,
        -2.83173195e-03,  1.19647038e-02,  0.00000000e+00],
       ...,
       [ 3.48726117e-01,  1.65733357e+00,  2.20498617e-02, ...,
         9.30666007e-04, -2.00520189e-03,  0.00000000e+00],
       [-1.57199576e-01, -1.18011709e+00,  2.88039260e-02, ...,
         2.17746360e-03, -2.50380372e-03,  0.00000000e+00],
       [-2.72943184e-01, -1.56685696e+00,  5.70091666e-02, ...,
         2.17746360e-03, -8.80324673e-03,  0.00000000e+00]])

.base_values =
array([6.79570243, 6.79570243, 6.79570243, ..., 6.79570243, 6.79570243,
       6.79570243])

.data =
array([['00-0031493', 'Jarryd Hayne', '2.0', ..., 1, 1, 1],
       ['00-0024559', 'D.J. Ware', '6.0', ..., 1, 1, 0],
       ['00-0036875', 'Rhamondre Stevenson', '16.0', ..., 0, 0, 0],
       ...,
       ['00-0034109', 'Phillip Lindsay', '7.0', ..., 1, 1, 1],
       ['00-0028087', 'Dion Lewis', '10.0', ..., 1, 1, 1],
       ['00-0027888', 'Toby Gerhart', '14.0', ..., 1, 1, 1]], dtype=object)

shap.plots.beeswarm(shap_values_rb, max_display = 25)

# Training split
te_train = te_eda_df.sample(n=round(len(te_eda_df)*.7), random_state=87)
te_train_X = te_train.loc[:, te_train.columns!='fantasy_points_ppr']
te_train_y = te_train['fantasy_points_ppr']
te_train_indices = te_train.index

te_nontrain = te_eda_df.drop(te_train.index)

# Test split
te_test = te_nontrain.sample(n=round(len(te_nontrain)*.5), random_state=87)
te_test_X = te_test.loc[:, te_test.columns!='fantasy_points_ppr']
te_test_y = te_test['fantasy_points_ppr']
te_test_X_indices = te_test_X.index

# Validation split
te_val = te_nontrain.drop(te_test.index)
te_val_X = te_val.loc[:, te_val.columns!='fantasy_points_ppr']
te_val_y = te_val['fantasy_points_ppr']
te_val

# Parameter grid
iterations_ls = [500, 1000]
learning_rates_ls = [0.01, 0.03, 0.1]
depth_list = [4,6,8]

te_cv_results = []

te_cvpool = Pool(data=te_train_X, label=te_train_y, cat_features=['player_id', 'player_name', 'position', 'team', 'opponent_team', 'day_slate', 'location'])

# Grid search
for i, l, d in tqdm(product(iterations_ls, learning_rates_ls, depth_list),
                    total = len(iterations_ls)*len(learning_rates_ls)*len(depth_list)):
        te_params = {'loss_function':'MAE',
                     'iterations': i,          
                     'learning_rate': l,       
                     'depth': d,
                     'early_stopping_rounds': 50,
                     'verbose': False, }                  
        te_cvscores = cv(pool=te_cvpool, params = te_params, fold_count = 5, shuffle=True, logging_level='Silent', partition_random_seed=30)
        best_te_cvscore = te_cvscores['test-MAE-mean'].min()
        te_cv_results.append({'iterations': i,
                              'learning_rate': l,
                              'depth': d,
                              'mae': best_te_cvscore})

te_cv_results_df = pd.DataFrame(te_cv_results)
best_params_te = te_cv_results_df.sort_values(by='mae').iloc[0]
best_params_te

100%|██████████| 18/18 [05:10<00:00, 17.23s/it]

iterations       1000.000000
learning_rate       0.010000
depth               6.000000
mae                 3.857413
Name: 10, dtype: float64

# Training Model on Full Training Set
final_te_params = {
    'loss_function':'MAE',
    'iterations': int(best_params_te['iterations']),
    'learning_rate': float(best_params_te['learning_rate']),
    'depth': int(best_params_te['depth']),
    'early_stopping_rounds': 50,
    'verbose': False
}

te_catboost = CatBoostRegressor(**final_te_params)
te_catboost.fit(te_train_X, te_train_y, cat_features=['player_id', 'player_name', 'position', 'team', 'opponent_team', 'day_slate', 'location'], 
                eval_set=(te_val_X, te_val_y))

<catboost.core.CatBoostRegressor at 0x130177820>

te_val_preds = te_catboost.predict(te_val_X)
print(f'Validation MAE: {mean_absolute_error(te_val_y, te_val_preds)}')
print(f'Validation R2: {r2_score(te_val_y, te_val_preds)}')

Validation MAE: 3.8972673833812124
Validation R2: 0.20043526074798168

te_test_preds = te_catboost.predict(te_test_X)
print(f'Test MAE: {mean_absolute_error(te_test_y, te_test_preds)}')
print(f'Test R2: {r2_score(te_test_y, te_test_preds)}')

# Predictions with indices
te_test_preds_wind = np.column_stack((te_test_X_indices, te_test_preds))
te_sorted_preds = te_test_preds_wind[te_test_preds_wind[:,0].argsort()]
te_sorted_preds_df = pd.DataFrame(te_sorted_preds, columns = ['index', 'prediction'])
te_sorted_preds_df.set_index('index', inplace=True)
te_sorted_preds_v_test = pd.concat([te_test_X, te_sorted_preds_df], axis = 1)
te_test_preds_exp = pd.concat([te_sorted_preds_v_test, te_test_y], axis=1)
te_test_preds_exp

Test MAE: 3.8118629894961376
Test R2: 0.1928275544515755

explainer_te = shap.TreeExplainer(te_catboost, feature_perturbation='tree_path_dependent')
shap_values_te = explainer_te(te_test_X)
shap_values_te

.values =
array([[-1.34097510e+00, -8.81559999e-01, -8.50072097e-04, ...,
        -4.18072525e-03, -4.58215485e-02,  0.00000000e+00],
       [ 1.03628328e+00,  7.10146367e-01, -9.24352913e-03, ...,
        -7.89735589e-03, -1.49020723e-01,  0.00000000e+00],
       [-1.04616212e+00, -7.27929274e-01,  1.80522028e-03, ...,
        -6.32297324e-03, -6.91807153e-02,  0.00000000e+00],
       ...,
       [ 2.17740840e+00,  1.61442279e+00,  1.31405546e-03, ...,
        -2.23466740e-02, -1.87429663e-01,  0.00000000e+00],
       [-1.23955879e+00, -6.80632689e-01,  1.64181763e-03, ...,
        -4.51309325e-03,  3.31184060e-02,  0.00000000e+00],
       [ 6.48646683e-01,  1.69732952e-01,  1.29791283e-03, ...,
         1.49564316e-02,  6.49123789e-02,  0.00000000e+00]])

.base_values =
array([5.07490613, 5.07490613, 5.07490613, ..., 5.07490613, 5.07490613,
       5.07490613])

.data =
array([['00-0027761', 'Brody Eldridge', '4.0', ..., 1, 1, 1],
       ['00-0027061', 'Jared Cook', '9.0', ..., 1, 1, 0],
       ['00-0028042', 'Luke Stocker', '3.0', ..., 1, 1, 1],
       ...,
       ['00-0022127', 'Jason Witten', '7.0', ..., 1, 1, 1],
       ['00-0034988', 'Kaden Smith', '14.0', ..., 1, 0, 0],
       ['00-0038996', 'Tucker Kraft', '4.0', ..., 0, 0, 0]], dtype=object)

shap.plots.beeswarm(shap_values_te, max_display = 25)

all_catboost_df = lagged_stats.copy()
all_catboost_df['season'] = all_catboost_df['season'].astype(str)
all_catboost_df['week'] = all_catboost_df['week'].astype(str)
all_catboost_df['game_number'] = all_catboost_df['game_number'].astype(str)

# Training split
all_train = all_catboost_df.sample(n=round(len(all_catboost_df)*.7), random_state=99)
all_train_X = all_train.loc[:, all_train.columns!='fantasy_points_ppr']
all_train_y = all_train['fantasy_points_ppr']
all_train_indices = all_train.index

all_nontrain = all_catboost_df.drop(all_train.index)

# test split
all_test = all_nontrain.sample(n=round(len(all_nontrain)*.5), random_state=99)
all_test_X = all_test.loc[:, all_test.columns!='fantasy_points_ppr']
all_test_y = all_test['fantasy_points_ppr']
all_test_X_indices = all_test_X.index

# Validation split
all_val = all_nontrain.drop(all_test.index)
all_val_X = all_val.loc[:, all_val.columns!='fantasy_points_ppr']
all_val_y = all_val['fantasy_points_ppr']
all_val

# Parameter grid
iterations_ls = [500, 1000]
learning_rates_ls = [0.01, 0.03, 0.1]
depth_list = [4,6,8]

all_cv_results = []

all_cvpool = Pool(data=all_train_X, label=all_train_y, cat_features=['player_id', 'player_name', 'position', 'team', 'opponent_team', 'day_slate', 'location'])

# Grid search
for i, l, d in tqdm(product(iterations_ls, learning_rates_ls, depth_list),
                    total = len(iterations_ls)*len(learning_rates_ls)*len(depth_list)):
        all_params = {'loss_function':'MAE',
                     'iterations': i,          
                     'learning_rate': l,       
                     'depth': d,
                     'early_stopping_rounds': 50,
                     'verbose': False, }                  
        all_cvscores = cv(pool=all_cvpool, params = all_params, fold_count = 5, shuffle=True, logging_level='Silent', partition_random_seed=30)
        best_all_cvscore = all_cvscores['test-MAE-mean'].min()
        all_cv_results.append({'iterations': i,
                              'learning_rate': l,
                              'depth': d,
                              'mae': best_all_cvscore})

all_cv_results_df = pd.DataFrame(all_cv_results)
best_params_all = all_cv_results_df.sort_values(by='mae').iloc[0]
best_params_all

100%|██████████| 18/18 [24:49<00:00, 82.78s/it]

iterations       1000.000000
learning_rate       0.030000
depth               6.000000
mae                 4.796267
Name: 13, dtype: float64

# Training Model on Full Training Set
final_all_params = {
    'loss_function':'MAE',
    'iterations': int(best_params_all['iterations']),
    'learning_rate': float(best_params_all['learning_rate']),
    'depth': int(best_params_all['depth']),
    'early_stopping_rounds': 50,
    'verbose': False
}

all_catboost = CatBoostRegressor(**final_all_params)
all_catboost.fit(all_train_X, all_train_y, cat_features=['player_id', 'player_name', 'position', 'team', 'opponent_team', 'day_slate', 'location'], 
                eval_set=(all_val_X, all_val_y))

<catboost.core.CatBoostRegressor at 0x130dfbf10>

all_val_preds = all_catboost.predict(all_val_X)
print(f'Validation MAE: {mean_absolute_error(all_val_y, all_val_preds)}')
print(f'Validation R2: {r2_score(all_val_y, all_val_preds)}')

Validation MAE: 4.760621630165713
Validation R2: 0.31058318805401275

all_test_preds = all_catboost.predict(all_test_X)
print(f'test MAE: {mean_absolute_error(all_test_y, all_test_preds)}')
print(f'test R2: {r2_score(all_test_y, all_test_preds)}')

# Predictions with indices
all_test_preds_wind = np.column_stack((all_test_X_indices, all_test_preds))
all_sorted_preds = all_test_preds_wind[all_test_preds_wind[:,0].argsort()]
all_sorted_preds_df = pd.DataFrame(all_sorted_preds, columns = ['index', 'prediction'])
all_sorted_preds_df.set_index('index', inplace=True)
all_sorted_preds_v_test = pd.concat([all_test_X, all_sorted_preds_df], axis = 1)
all_test_preds_exp = pd.concat([all_sorted_preds_v_test, all_test_y], axis=1)
all_test_preds_exp

test MAE: 4.810515250800812
test R2: 0.3193426864681951

explainer_all = shap.TreeExplainer(all_catboost, feature_perturbation='tree_path_dependent')
shap_values_all = explainer_all(all_test_X)
shap_values_all

.values =
array([[ 1.13992813e-01, -2.09576267e+00, -7.85204658e-03, ...,
        -1.57291643e-02,  4.13812665e-02, -3.35678138e-04],
       [-7.15732720e-01, -1.26657054e+00, -1.91032380e-02, ...,
         1.41020834e-03,  1.32839234e-02, -3.35678138e-04],
       [ 1.25002261e+00,  3.66298376e+00, -1.03755686e-01, ...,
        -2.02982350e-02,  6.79289673e-02, -3.35678138e-04],
       ...,
       [-3.68746538e-01, -2.12202149e+00, -5.15444101e-02, ...,
        -6.25791788e-03, -4.46963980e-02, -1.03321552e-03],
       [ 8.34025070e-01,  3.62403903e+00,  3.21448225e-02, ...,
         4.28819542e-03,  6.02918386e-02,  2.76964966e-04],
       [-3.43979700e-01, -1.91570773e+00, -3.90298041e-03, ...,
        -5.73140218e-03, -6.69356767e-02, -3.35678138e-04]])

.base_values =
array([7.36996025, 7.36996025, 7.36996025, ..., 7.36996025, 7.36996025,
       7.36996025])

.data =
array([['00-0029141', 'Alfred Morris', '4.0', ..., 0, 0, 0],
       ['00-0025463', 'Jason Hill', '10.0', ..., 1, 1, 0],
       ['00-0036212', 'Tua Tagovailoa', '2.0', ..., 0, 0, 0],
       ...,
       ['00-0026286', 'Jerome Felton', '1.0', ..., 1, 1, 0],
       ['00-0030496', "Le'Veon Bell", '6.0', ..., 1, 0, 0],
       ['00-0034279', 'Equanimeous St. Brown', '3.0', ..., 1, 1, 0]],
      dtype=object)

shap.plots.beeswarm(shap_values_all, max_display = 25)

by_season = lagged_stats.groupby(['player_name', 'season']).agg(
    {'fantasy_points_ppr':'sum',
     'never_seen':'mean',
     'DNP_prior_season_game':'mean'}
).reset_index()
by_season.describe()
## 25th percentile of target variable = ~11 points, so do QBs with 40+ points in a fantasy season only

all_filter_df = by_season[(by_season['fantasy_points_ppr'] >= 11) & (by_season['never_seen'] != 1) & (by_season['DNP_prior_season_game'] != 1)]
all_filtered = lagged_stats.merge(all_filter_df[['player_name', 'season']], how='inner', on=['player_name', 'season'])
all_filtered

# Training split
allf_train = all_filtered.sample(n=round(len(all_filtered)*.7), random_state=5)
allf_train_X = allf_train.loc[:, allf_train.columns!='fantasy_points_ppr']
allf_train_y = allf_train['fantasy_points_ppr']
allf_train_indices = allf_train.index

allf_nontrain = all_filtered.drop(allf_train.index)

# Test split
allf_test = allf_nontrain.sample(n=round(len(allf_nontrain)*.5), random_state=5)
allf_test_X = allf_test.loc[:, allf_test.columns!='fantasy_points_ppr']
allf_test_y = allf_test['fantasy_points_ppr']
allf_test_X_indices = allf_test_X.index

# Validation split
allf_val = allf_nontrain.drop(allf_test.index)
allf_val_X = allf_val.loc[:, allf_val.columns!='fantasy_points_ppr']
allf_val_y = allf_val['fantasy_points_ppr']
allf_val

# Parameter grid
iterations_ls = [500, 1000]
learning_rates_ls = [0.01, 0.03, 0.1]
depth_list = [4,6,8]

allf_cv_results = []

allf_cvpool = Pool(data=allf_train_X, label=allf_train_y, cat_features=['player_id', 'player_name', 'position', 'team', 'opponent_team', 'day_slate', 'location'])

# Grid search
for i, l, d in tqdm(product(iterations_ls, learning_rates_ls, depth_list),
                    total = len(iterations_ls)*len(learning_rates_ls)*len(depth_list)):
        all_params = {'loss_function':'MAE',
                     'iterations': i,          
                     'learning_rate': l,       
                     'depth': d,
                     'early_stopping_rounds': 50,
                     'verbose': False, }                  
        allf_cvscores = cv(pool=allf_cvpool, params = all_params, fold_count = 5, shuffle=True, logging_level='Silent', partition_random_seed=5)
        best_allf_cvscore = allf_cvscores['test-MAE-mean'].min()
        allf_cv_results.append({'iterations': i,
                              'learning_rate': l,
                              'depth': d,
                              'mae': best_allf_cvscore})

allf_cv_results_df = pd.DataFrame(allf_cv_results)
best_params_allf = allf_cv_results_df.sort_values(by='mae').iloc[0]
best_params_allf

100%|██████████| 18/18 [21:05<00:00, 70.28s/it]

iterations       1000.00000
learning_rate       0.03000
depth               6.00000
mae                 5.07422
Name: 13, dtype: float64

# Training Model on Full Training Set
final_allf_params = {
    'loss_function':'MAE',
    'iterations': int(best_params_allf['iterations']),
    'learning_rate': float(best_params_allf['learning_rate']),
    'depth': int(best_params_allf['depth']),
    'early_stopping_rounds': 50,
    'verbose': False
}

allf_catboost = CatBoostRegressor(**final_allf_params)
allf_catboost.fit(allf_train_X, allf_train_y, cat_features=['player_id', 'player_name', 'position', 'team', 'opponent_team', 'day_slate', 'location'], 
                eval_set=(allf_val_X, allf_val_y))

<catboost.core.CatBoostRegressor at 0x130173790>

allf_val_preds = allf_catboost.predict(allf_val_X)
print(f'Validation MAE: {mean_absolute_error(allf_val_y, allf_val_preds)}')
print(f'Validation R2: {r2_score(allf_val_y, allf_val_preds)}')

Validation MAE: 5.131577748906053
Validation R2: 0.27700435434086523

allf_test_preds = allf_catboost.predict(allf_test_X)
print(f'Test MAE: {mean_absolute_error(allf_test_y, allf_test_preds)}')
print(f'Test R2: {r2_score(allf_test_y, allf_test_preds)}')

# Predictions with indices
allf_test_preds_wind = np.column_stack((allf_test_X_indices, allf_test_preds))
allf_sorted_preds = allf_test_preds_wind[allf_test_preds_wind[:,0].argsort()]
allf_sorted_preds_df = pd.DataFrame(allf_sorted_preds, columns = ['index', 'prediction'])
allf_sorted_preds_df.set_index('index', inplace=True)
allf_sorted_preds_v_test = pd.concat([allf_test_X, allf_sorted_preds_df], axis = 1)
allf_test_preds_exp = pd.concat([allf_sorted_preds_v_test, allf_test_y], axis=1)
allf_test_preds_exp

Test MAE: 5.1044939256398
Test R2: 0.2773448827651671

from sklearn.preprocessing import LabelEncoder, OrdinalEncoder

deep_lagged_stats = lagged_stats.copy()
deep_lagged_stats.drop(columns=['player_name'], inplace=True)

# Getting position dummies
deep_lagged_stats = pd.get_dummies(deep_lagged_stats, columns=['position'], prefix = 'cat')
deep_lagged_stats[['cat_QB', 'cat_WR', 'cat_RB', 'cat_FB', 'cat_TE']] = deep_lagged_stats[['cat_QB', 'cat_WR', 'cat_RB', 'cat_FB', 'cat_TE']].astype(int)

# Label Encoding
game_number_encoder = LabelEncoder()
team_encoder = LabelEncoder()
week_encoder = LabelEncoder()
opponent_team_encoder = LabelEncoder()

# Ordinal Encoding - variables that will have previously unseen values (season, player_id, and location)
ordinal_encoder = OrdinalEncoder(
    handle_unknown = 'use_encoded_value',
    unknown_value=-1
)

deep_lagged_stats['game_number_encoded'] = game_number_encoder.fit_transform(deep_lagged_stats['game_number'])
deep_lagged_stats['team_encoded'] = team_encoder.fit_transform(deep_lagged_stats['team'])
deep_lagged_stats['week_encoded'] = week_encoder.fit_transform(deep_lagged_stats['week'])
deep_lagged_stats['opponent_team_encoded'] = opponent_team_encoder.fit_transform(deep_lagged_stats['opponent_team'])
deep_lagged_stats[['player_id_encoded', 'season_encoded', 'location_encoded', 'day_slate_encoded']] = ordinal_encoder.fit_transform(deep_lagged_stats[['player_id', 'season', 'location', 'day_slate']])
deep_lagged_stats[['player_id_encoded', 'season_encoded', 'location_encoded', 'day_slate_encoded']] += 1

# Get number of unique values for each categorical variable
num_player_id = deep_lagged_stats['player_id_encoded'].nunique()
print(f'Number of unique player_id: {num_player_id}')
num_game_number = deep_lagged_stats['game_number_encoded'].nunique()
print(f'Number of unique game_number: {num_game_number}')
num_team = deep_lagged_stats['team_encoded'].nunique()
print(f'Number of unique team: {num_team}')
num_season = deep_lagged_stats['season_encoded'].nunique()
print(f'Number of unique season: {num_season}')
num_week = deep_lagged_stats['week_encoded'].nunique()
print(f'Number of unique week: {num_week}')
num_opponent_team = deep_lagged_stats['opponent_team_encoded'].nunique()
print(f'Number of unique opponent_team: {num_opponent_team}')
num_day_slate = deep_lagged_stats['day_slate_encoded'].nunique()
print(f'Number of unique day_slate: {num_day_slate}')
num_location = deep_lagged_stats['location_encoded'].nunique()
print(f'Number of unique location: {num_location}')

# Prepare features and target
X_deep = deep_lagged_stats.drop(columns=['fantasy_points_ppr'])
y_deep = deep_lagged_stats['fantasy_points_ppr']
X_deep_training, X_deep_test, y_deep_training, y_deep_test = train_test_split(X_deep, y_deep, test_size = 0.2, random_state = 9)
X_deep_training.fillna(0, inplace=True)
y_deep_training.fillna(0, inplace=True)
X_deep_test.fillna(0, inplace=True)
y_deep_test.fillna(0, inplace=True)

Number of unique player_id: 2247
Number of unique game_number: 16
Number of unique team: 32
Number of unique season: 15
Number of unique week: 17
Number of unique opponent_team: 32
Number of unique day_slate: 22
Number of unique location: 33

pip install tensorflow

Requirement already satisfied: tensorflow in ./myenv3.10/lib/python3.10/site-packages (2.16.2)
Requirement already satisfied: google-pasta>=0.1.1 in ./myenv3.10/lib/python3.10/site-packages (from tensorflow) (0.2.0)
Requirement already satisfied: packaging in ./myenv3.10/lib/python3.10/site-packages (from tensorflow) (25.0)
Requirement already satisfied: flatbuffers>=23.5.26 in ./myenv3.10/lib/python3.10/site-packages (from tensorflow) (25.2.10)
Requirement already satisfied: h5py>=3.10.0 in ./myenv3.10/lib/python3.10/site-packages (from tensorflow) (3.14.0)
Requirement already satisfied: opt-einsum>=2.3.2 in ./myenv3.10/lib/python3.10/site-packages (from tensorflow) (3.4.0)
Requirement already satisfied: grpcio<2.0,>=1.24.3 in ./myenv3.10/lib/python3.10/site-packages (from tensorflow) (1.73.0)
Requirement already satisfied: six>=1.12.0 in ./myenv3.10/lib/python3.10/site-packages (from tensorflow) (1.17.0)
Requirement already satisfied: termcolor>=1.1.0 in ./myenv3.10/lib/python3.10/site-packages (from tensorflow) (3.1.0)
Requirement already satisfied: libclang>=13.0.0 in ./myenv3.10/lib/python3.10/site-packages (from tensorflow) (18.1.1)
Requirement already satisfied: protobuf!=4.21.0,!=4.21.1,!=4.21.2,!=4.21.3,!=4.21.4,!=4.21.5,<5.0.0dev,>=3.20.3 in ./myenv3.10/lib/python3.10/site-packages (from tensorflow) (4.25.8)
Requirement already satisfied: astunparse>=1.6.0 in ./myenv3.10/lib/python3.10/site-packages (from tensorflow) (1.6.3)
Requirement already satisfied: gast!=0.5.0,!=0.5.1,!=0.5.2,>=0.2.1 in ./myenv3.10/lib/python3.10/site-packages (from tensorflow) (0.6.0)
Requirement already satisfied: setuptools in ./myenv3.10/lib/python3.10/site-packages (from tensorflow) (63.2.0)
Requirement already satisfied: numpy<2.0.0,>=1.23.5 in ./myenv3.10/lib/python3.10/site-packages (from tensorflow) (1.26.4)
Requirement already satisfied: tensorboard<2.17,>=2.16 in ./myenv3.10/lib/python3.10/site-packages (from tensorflow) (2.16.2)
Requirement already satisfied: wrapt>=1.11.0 in ./myenv3.10/lib/python3.10/site-packages (from tensorflow) (1.17.2)
Requirement already satisfied: requests<3,>=2.21.0 in ./myenv3.10/lib/python3.10/site-packages (from tensorflow) (2.32.4)
Requirement already satisfied: tensorflow-io-gcs-filesystem>=0.23.1 in ./myenv3.10/lib/python3.10/site-packages (from tensorflow) (0.37.1)
Requirement already satisfied: keras>=3.0.0 in ./myenv3.10/lib/python3.10/site-packages (from tensorflow) (3.10.0)
Requirement already satisfied: absl-py>=1.0.0 in ./myenv3.10/lib/python3.10/site-packages (from tensorflow) (2.3.0)
Requirement already satisfied: typing-extensions>=3.6.6 in ./myenv3.10/lib/python3.10/site-packages (from tensorflow) (4.14.0)
Requirement already satisfied: ml-dtypes~=0.3.1 in ./myenv3.10/lib/python3.10/site-packages (from tensorflow) (0.3.2)
Requirement already satisfied: wheel<1.0,>=0.23.0 in ./myenv3.10/lib/python3.10/site-packages (from astunparse>=1.6.0->tensorflow) (0.45.1)
Requirement already satisfied: namex in ./myenv3.10/lib/python3.10/site-packages (from keras>=3.0.0->tensorflow) (0.1.0)
Requirement already satisfied: rich in ./myenv3.10/lib/python3.10/site-packages (from keras>=3.0.0->tensorflow) (14.0.0)
Requirement already satisfied: optree in ./myenv3.10/lib/python3.10/site-packages (from keras>=3.0.0->tensorflow) (0.16.0)
Requirement already satisfied: certifi>=2017.4.17 in ./myenv3.10/lib/python3.10/site-packages (from requests<3,>=2.21.0->tensorflow) (2025.4.26)
Requirement already satisfied: urllib3<3,>=1.21.1 in ./myenv3.10/lib/python3.10/site-packages (from requests<3,>=2.21.0->tensorflow) (2.4.0)
Requirement already satisfied: idna<4,>=2.5 in ./myenv3.10/lib/python3.10/site-packages (from requests<3,>=2.21.0->tensorflow) (3.10)
Requirement already satisfied: charset_normalizer<4,>=2 in ./myenv3.10/lib/python3.10/site-packages (from requests<3,>=2.21.0->tensorflow) (3.4.2)
Requirement already satisfied: tensorboard-data-server<0.8.0,>=0.7.0 in ./myenv3.10/lib/python3.10/site-packages (from tensorboard<2.17,>=2.16->tensorflow) (0.7.2)
Requirement already satisfied: markdown>=2.6.8 in ./myenv3.10/lib/python3.10/site-packages (from tensorboard<2.17,>=2.16->tensorflow) (3.8)
Requirement already satisfied: werkzeug>=1.0.1 in ./myenv3.10/lib/python3.10/site-packages (from tensorboard<2.17,>=2.16->tensorflow) (3.1.3)
Requirement already satisfied: MarkupSafe>=2.1.1 in ./myenv3.10/lib/python3.10/site-packages (from werkzeug>=1.0.1->tensorboard<2.17,>=2.16->tensorflow) (3.0.2)
Requirement already satisfied: pygments<3.0.0,>=2.13.0 in ./myenv3.10/lib/python3.10/site-packages (from rich->keras>=3.0.0->tensorflow) (2.19.1)
Requirement already satisfied: markdown-it-py>=2.2.0 in ./myenv3.10/lib/python3.10/site-packages (from rich->keras>=3.0.0->tensorflow) (3.0.0)
Requirement already satisfied: mdurl~=0.1 in ./myenv3.10/lib/python3.10/site-packages (from markdown-it-py>=2.2.0->rich->keras>=3.0.0->tensorflow) (0.1.2)

[notice] A new release of pip available: 22.2.2 -> 25.1.1
[notice] To update, run: pip install --upgrade pip
Note: you may need to restart the kernel to use updated packages.

from tensorflow.keras import Input, Model
from tensorflow.keras.layers import Embedding, Flatten, Concatenate, Dense
import tensorflow.keras.layers as layers  # use this consistently
from tensorflow.keras.regularizers import l1_l2

# Define output_dim values
print(f'Rule of Thumb player_id: {min(50, round(num_player_id)**0.25)}')
embedding_dim_player_id = 16
embedding_dim_game_number = 4
embedding_dim_team = 4
embedding_dim_season = 4
embedding_dim_week = 4
embedding_dim_opponent_team = 4
embedding_dim_day_slate = 4
embedding_dim_location = 4

# Input layers
player_id_input = Input(shape=(1,), name='player_id_input')
game_number_input = Input(shape=(1,), name='game_number_input')
team_input = Input(shape=(1,), name='team_input')
season_input = Input(shape=(1,), name='season_input')
week_input = Input(shape=(1,), name='week_input')
opponent_team_input = Input(shape=(1,), name='opponent_team_input')
day_slate_input = Input(shape=(1,), name='day_slate_input')
location_input = Input(shape=(1,), name='location_input')
numeric_inputs = Input(shape=(50,), name='numeric_input') # shape is all numeric columns besides encoded columns

# Embedding layers applied to inputs
player_id_emb = Flatten()(Embedding(input_dim=num_player_id + 2, output_dim=embedding_dim_player_id)(player_id_input))
game_number_emb = Flatten()(Embedding(input_dim=num_game_number + 1, output_dim=embedding_dim_game_number)(game_number_input))
team_emb = Flatten()(Embedding(input_dim=num_team + 1, output_dim=embedding_dim_team)(team_input))
season_emb = Flatten()(Embedding(input_dim=num_season + 2, output_dim=embedding_dim_season)(season_input))
week_emb = Flatten()(Embedding(input_dim=num_week + 1, output_dim=embedding_dim_week)(week_input))
opponent_team_emb = Flatten()(Embedding(input_dim=num_opponent_team + 1, output_dim=embedding_dim_opponent_team)(opponent_team_input))
day_slate_emb = Flatten()(Embedding(input_dim=num_day_slate + 2, output_dim=embedding_dim_day_slate)(day_slate_input))
location_emb = Flatten()(Embedding(input_dim=num_location + 2, output_dim=embedding_dim_location)(location_input))

# Concatenate all features
concatenated = Concatenate()([
    player_id_emb,
    game_number_emb,
    team_emb,
    season_emb,
    week_emb,
    opponent_team_emb,
    day_slate_emb,
    location_emb,
    numeric_inputs
])

# Dense layers
## Originally had 3 layers, but that caused overfitting
x = Dense(128, activation='relu', kernel_regularizer = l1_l2(l1=0.000001, l2=0.001))(concatenated) # Added in elastic net regularization with more powerful L2 than L1 to preserve features more
x = Dense(64, activation='relu', kernel_regularizer = l1_l2(l1=0.000001, l2=0.001))(x)
output = Dense(1)(x)

# Build model
deep_model = Model(inputs=[
    player_id_input, game_number_input, team_input, season_input,
    week_input, opponent_team_input, day_slate_input, location_input,
    numeric_inputs
], outputs=output)

deep_model.compile(optimizer='adam', loss='mean_absolute_error', metrics=['mae']) # adam optimizer is most robust for regression tasks

# Prepare the input data
player_id_input_train = X_deep_training['player_id_encoded'].values
game_number_input_train = X_deep_training['game_number_encoded'].values
team_input_train = X_deep_training['team_encoded'].values
season_input_train = X_deep_training['season_encoded'].values
week_input_train = X_deep_training['week_encoded'].values
opponent_team_input_train = X_deep_training['opponent_team_encoded'].values
day_slate_input_train = X_deep_training['day_slate_encoded'].values
location_input_train = X_deep_training['location_encoded'].values
numeric_features_train = X_deep_training[['completions_lagged', 'attempts_lagged', 'passing_yards_lagged',
       'passing_tds_lagged', 'interceptions_lagged', 'sacks_lagged',
       'sack_yards_lagged', 'sack_fumbles_lagged', 'sack_fumbles_lost_lagged',
       'passing_air_yards_lagged', 'passing_yards_after_catch_lagged',
       'passing_first_downs_lagged', 'passing_epa_lagged',
       'passing_2pt_conversions_lagged', 'pacr_lagged', 'dakota_lagged',
       'carries_lagged', 'rushing_yards_lagged', 'rushing_tds_lagged',
       'rushing_fumbles_lagged', 'rushing_fumbles_lost_lagged',
       'rushing_first_downs_lagged', 'rushing_epa_lagged',
       'rushing_2pt_conversions_lagged', 'receptions_lagged', 'targets_lagged',
       'receiving_yards_lagged', 'receiving_tds_lagged',
       'receiving_fumbles_lagged', 'receiving_fumbles_lost_lagged',
       'receiving_air_yards_lagged', 'receiving_yards_after_catch_lagged',
       'receiving_first_downs_lagged', 'receiving_epa_lagged',
       'receiving_2pt_conversions_lagged', 'racr_lagged',
       'target_share_lagged', 'air_yards_share_lagged', 'wopr_lagged',
       'special_teams_tds_lagged', 'fantasy_points_ppr_lagged',
       'used_opponent_avg', 'used_game_number_avg', 'DNP_prior_season_game',
       'never_seen', 'cat_FB', 'cat_QB', 'cat_RB', 'cat_TE', 'cat_WR']].values

# Train the model
deep_model.fit([
    player_id_input_train, game_number_input_train, team_input_train, season_input_train, week_input_train, opponent_team_input_train, day_slate_input_train, location_input_train,
    numeric_features_train],
    y_deep_training,
    epochs = 50,
    batch_size=32,
    validation_split=0.2,
    verbose=False
    )

# Evaluate on training set
train_loss, train_mae = deep_model.evaluate([
    player_id_input_train, game_number_input_train, team_input_train, season_input_train,
    week_input_train, opponent_team_input_train, day_slate_input_train, location_input_train,
    numeric_features_train
], y_deep_training, verbose=0)
print(f"Train MAE: {train_mae:.4f}")

# Evaluate the model
player_id_input_test = X_deep_test['player_id_encoded'].values
game_number_input_test = X_deep_test['game_number_encoded'].values
team_input_test = X_deep_test['team_encoded'].values
season_input_test = X_deep_test['season_encoded'].values
week_input_test = X_deep_test['week_encoded'].values
opponent_team_input_test = X_deep_test['opponent_team_encoded'].values
day_slate_input_test = X_deep_test['day_slate_encoded'].values
location_input_test = X_deep_test['location_encoded'].values
numeric_features_test = X_deep_test[['completions_lagged', 'attempts_lagged', 'passing_yards_lagged',
       'passing_tds_lagged', 'interceptions_lagged', 'sacks_lagged',
       'sack_yards_lagged', 'sack_fumbles_lagged', 'sack_fumbles_lost_lagged',
       'passing_air_yards_lagged', 'passing_yards_after_catch_lagged',
       'passing_first_downs_lagged', 'passing_epa_lagged',
       'passing_2pt_conversions_lagged', 'pacr_lagged', 'dakota_lagged',
       'carries_lagged', 'rushing_yards_lagged', 'rushing_tds_lagged',
       'rushing_fumbles_lagged', 'rushing_fumbles_lost_lagged',
       'rushing_first_downs_lagged', 'rushing_epa_lagged',
       'rushing_2pt_conversions_lagged', 'receptions_lagged', 'targets_lagged',
       'receiving_yards_lagged', 'receiving_tds_lagged',
       'receiving_fumbles_lagged', 'receiving_fumbles_lost_lagged',
       'receiving_air_yards_lagged', 'receiving_yards_after_catch_lagged',
       'receiving_first_downs_lagged', 'receiving_epa_lagged',
       'receiving_2pt_conversions_lagged', 'racr_lagged',
       'target_share_lagged', 'air_yards_share_lagged', 'wopr_lagged',
       'special_teams_tds_lagged', 'fantasy_points_ppr_lagged',
       'used_opponent_avg', 'used_game_number_avg', 'DNP_prior_season_game',
       'never_seen', 'cat_FB', 'cat_QB', 'cat_RB', 'cat_TE', 'cat_WR']].values

test_loss, test_mae = deep_model.evaluate([
    player_id_input_test, game_number_input_test, team_input_test, season_input_test, week_input_test, opponent_team_input_test, day_slate_input_test, location_input_test,
    numeric_features_test], y_deep_test)
print(f"Test MAE: {test_mae:.4f}")

# Make predictions
predictions = deep_model.predict([
    player_id_input_test, game_number_input_test, team_input_test, season_input_test, week_input_test, opponent_team_input_test, day_slate_input_test, location_input_test,
    numeric_features_test])
predictions

2025-06-24 11:55:41.069680: I tensorflow/core/platform/cpu_feature_guard.cc:210] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.
To enable the following instructions: AVX2 AVX512F AVX512_VNNI FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.

Rule of Thumb player_id: 6.8849496423696035
Train MAE: 4.5069
450/450 ━━━━━━━━━━━━━━━━━━━━ 1s 2ms/step - loss: 5.0012 - mae: 4.9574
Test MAE: 4.8846
450/450 ━━━━━━━━━━━━━━━━━━━━ 1s 2ms/step

array([[ 8.431002 ],
       [ 3.9771852],
       [ 5.788801 ],
       ...,
       [ 1.8462148],
       [ 7.87333  ],
       [16.727718 ]], dtype=float32)

roster25 = nfl.import_seasonal_rosters([2025])[['player_id', 'player_name', 'season', 'team', 'depth_chart_position', 'entry_year', 'rookie_year']]
roster25 = roster25[(roster25['depth_chart_position'].isin(['QB', 'TE', 'RB', 'FB', 'WR'])) & (roster25['entry_year'] != 2025) & (roster25['rookie_year'] != 2025)]
roster25['season'] = roster25['season'].astype(str)
roster25.rename(columns={'depth_chart_position':'position'}, inplace=True)
roster25

schedule25 = pd.read_csv('NFL Schedule 2025 - Sheet1.csv')
schedule25 = schedule25.iloc[1:].reset_index(drop=True)

# Creating team_abbreviation mapping
team_abbrev = {
    'Philadelphia Eagles':'PHI', 'Los Angeles Chargers': 'LAC', 'Atlanta Falcons': 'ATL', 'Cleveland Browns': 'CLE', 'Indianapolis Colts': 'IND',
    'Jacksonville Jaguars': 'JAX', 'New England Patriots': 'NE', 'New Orleans Saints': 'NO', 'New York Jets': 'NYJ', 'Washington Commanders': 'WAS',
    'Denver Broncos': 'DEN', 'Seattle Seahawks': 'SEA', 'Green Bay Packers': 'GB', 'Los Angeles Rams': 'LA', 'Buffalo Bills': 'BUF', 'Chicago Bears': 'CHI',
    'Baltimore Ravens': 'BAL', 'Cincinnati Bengals': 'CIN', 'Dallas Cowboys': 'DAL', 'Detroit Lions': 'DET', 'Miami Dolphins': 'MIA', 'Pittsburgh Steelers': 'PIT',
    'Tennessee Titans': 'TEN', 'Arizona Cardinals': 'ARI', 'Kansas City Chiefs': 'KC', 'Minnesota Vikings': 'MIN', 'Houston Texans': 'HOU',
    'Las Vegas Raiders': 'LV', 'Carolina Panthers': 'CAR', 'Tampa Bay Buccaneers': 'TB', 'San Francisco 49ers': 'SF', 'New York Giants': 'NYG'
}

# Cleaning messy old file
schedule25['home_team'] = schedule25['Unnamed: 4'].map(team_abbrev)
schedule25['away_team'] = schedule25['Unnamed: 5'].map(team_abbrev)
schedule25['location'] = np.where(
    schedule25['Unnamed: 3'].isin(['Wembley Stadium', 'Santiago Bernabeu Stadium', 'Tottenham Hotspur Stadium',
                                  'Arena Corinthians', 'Croke Park', 'Olympic Stadium']), 'global', schedule25['home_team'])

# Adding season column
schedule25['season'] = str(2025)

# Adding day_slate column 
schedule25['time'] = pd.to_datetime(schedule25['Unnamed: 6'])
schedule25['weekday'] = schedule25['time'].dt.day_name()
schedule25['hour'] = schedule25['time'].dt.hour
schedule25['slate'] = np.where(
    schedule25['hour'].isin([12, 13, 14]),
    'Morning',
    np.where(
        schedule25['hour'].isin([15]),
        'Noon',
        np.where(
            schedule25['hour'].isin([16, 17, 18, 19]),
            'Afternoon',
            np.where(
                schedule25['hour'].isin([20, 21, 22, 23]),
                'Night',
                'global'
            )
        )
    )
)
schedule25['day_slate'] = schedule25['weekday'] + '_' + schedule25['slate']
schedule25['week'] = schedule25['Unnamed: 1'].astype(str)
schedule25

# Formatting schedule df to merge with roster df
## Creating home dataframe for home teams
home25 = schedule25.copy()
home25['team'] = home25['home_team']
home25['opponent_team'] = home25['away_team']
home25['is_home'] = True

## Creating away dataframe for away teams
away25 = schedule25.copy()
away25['team'] = away25['away_team']
away25['opponent_team'] = away25['home_team']
away25['is_away'] = True

## Joining home and away dataframes to be able to merge with roster df
schedule25_long = pd.concat([home25, away25], ignore_index=True)
schedule25_long[schedule25_long['opponent_team'].isna()]
roster_schedule25 = pd.merge(roster25.drop(columns=['entry_year', 'rookie_year']), schedule25_long[['team', 'season','week', 'opponent_team', 'day_slate', 'location']], 
                             on=['season', 'team'], how='inner')

# game_number column creation
gn_rs25 = roster_schedule25.copy()
gn_rs25['week'] = gn_rs25['week'].astype(int)
game_no = gn_rs25[['season', 'week', 'team', 'location']].groupby(['season', 'team', 'week']).agg({'location': 'first'})
game_no['game_number'] = game_no.groupby(['season', 'team']).cumcount() + 1
game_no.reset_index(inplace=True)
game_no['week'] = game_no['week'].astype(str)

roster_schedule25 = roster_schedule25.merge(game_no, how='left', on=['season', 'team', 'location', 'week'])
roster_schedule25

players25 = weekly_stats[weekly_stats['season'] == 2024]
players25.rename(columns={'player_display_name':'player_name'}, inplace=True)
players25

/var/folders/7t/jxkc73q943n1l8zb9xmqg9kc0000gn/T/ipykernel_1185/3418235261.py:2: SettingWithCopyWarning:


A value is trying to be set on a copy of a slice from a DataFrame

See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy

# Average 2024 stats by player + opponent
player_vs_opp_avg = (
    players25
    .groupby(['player_id', 'player_name', 'opponent_team'])
    .mean(numeric_only=True)  # restrict to numeric columns (stats)
    .reset_index()
)

# Merge 2025 schedule with opponent-based stats
merged = roster_schedule25.merge(
    player_vs_opp_avg,
    how='left',
    on=['player_id', 'player_name', 'opponent_team'],
    suffixes=('', '_opp_avg')
)
merged.drop(columns=['game_number_opp_avg', 'season_opp_avg', 'week_opp_avg'], inplace=True)

# Identify where opponent-based stats are missing
missing_mask = merged['fantasy_points_ppr'].isna()

# Get fallback stats by game_number
fallback_stats = players25.copy()
merged_fallback = roster_schedule25[missing_mask].merge(
    fallback_stats,
    how='left',
    on=['player_id', 'player_name', 'game_number'],
    suffixes=('', '_fallback')
)
merged_fallback.drop(columns=['position_fallback', 'team_fallback', 'season_fallback', 
                              'week_fallback', 'opponent_team_fallback', 'day_slate_fallback', 'location_fallback'], inplace=True)

# Combine both: start with merged (has opponent-based stats), fill in missing with fallback
final_scoring = merged.copy()
# Align indices to ensure correct assignment
merged_fallback.index = final_scoring[missing_mask].index
for col in [c for c in players25.columns if c not in ['player_id', 'player_name', 'game_number']]:
    if col in final_scoring.columns and col in merged_fallback.columns:
        final_scoring.loc[missing_mask, col] = merged_fallback[col]

# Create indicator columns
final_scoring['used_opponent_avg'] = (~missing_mask).astype(int)
final_scoring['used_game_number_avg'] = (missing_mask).astype(int)

# Rename all stat columns to end in _lagged
rename_dict = {col: f"{col}_lagged" for col in stat_cols if col in final_scoring.columns}
final_scoring.rename(columns=rename_dict, inplace=True)

# Identify DNPs where no lagged stat data exists
final_scoring['DNP_prior_season_game'] = final_scoring[f"{stat_cols[0]}_lagged"].isna().astype(int)

# Compute prior season averages for each player
prior_season_avg = (
    players25
    .groupby('player_id')[stat_cols]
    .mean()
    .reset_index()
    .rename(columns={col: f"{col}_lag_avg_season" for col in stat_cols})
)

# Create lookup for players missing lagged stats
season_avg_df = final_scoring.loc[final_scoring['DNP_prior_season_game'] == 1, ['player_id']].copy()
season_avg_df = season_avg_df.merge(prior_season_avg, how='left', on='player_id')

# Fill missing values from season average
for col in stat_cols:
    col_lagged = f"{col}_lagged"
    col_avg = f"{col}_lag_avg_season"
    final_scoring.loc[final_scoring['DNP_prior_season_game'] == 1, col_lagged] = season_avg_df[col_avg].values

# Identify players who appeared in prior season data
seen_players = set(players25['player_id'].unique())

# Create never_seen column based on whether player_id was in 2024 data
final_scoring['never_seen'] = final_scoring['player_id'].apply(lambda pid: 0 if pid in seen_players else 1)
final_scoring.drop(columns=['fantasy_points_lagged'], inplace=True)
final_scoring['game_number'] = final_scoring['game_number'].astype(float)
final_scoring['game_number'] = final_scoring['game_number'].astype(str)
final_scoring = final_scoring[final_scoring['game_number'] != '17.0']
final_scoring

common_cols = [col for col in all_test_X.columns if col in final_scoring.columns]
final_scoring = final_scoring[common_cols]
final_scoring_indices = final_scoring.index
all_25_preds = all_catboost.predict(final_scoring)

all_25_catboost = final_scoring.copy()
all_25_catboost['prediction'] = all_25_preds
season_predictions25 = all_25_catboost.groupby(['player_name', 'season']).sum().reset_index()
season_predictions25

/var/folders/7t/jxkc73q943n1l8zb9xmqg9kc0000gn/T/ipykernel_1185/14465142.py:8: FutureWarning:

The default value of numeric_only in DataFrameGroupBy.sum is deprecated. In a future version, numeric_only will default to False. Either specify numeric_only or select only columns which should be valid for the function.

deep_final_scoring = final_scoring.copy()
deep_final_scoring['game_number'] = deep_final_scoring['game_number'].astype(float)
deep_final_scoring['game_number'] = deep_final_scoring['game_number'].astype(str)
deep_final_scoring.drop(columns=['player_name'], inplace=True)


# Getting position dummies
deep_final_scoring = pd.get_dummies(deep_final_scoring, columns=['position'], prefix = 'cat')
deep_final_scoring[['cat_QB', 'cat_WR', 'cat_RB', 'cat_FB', 'cat_TE']] = deep_final_scoring[['cat_QB', 'cat_WR', 'cat_RB', 'cat_FB', 'cat_TE']].astype(int)

deep_final_scoring['game_number_encoded'] = game_number_encoder.transform(deep_final_scoring['game_number'])
deep_final_scoring['team_encoded'] = team_encoder.transform(deep_final_scoring['team'])
deep_final_scoring['week_encoded'] = week_encoder.transform(deep_final_scoring['week'])
deep_final_scoring['opponent_team_encoded'] = opponent_team_encoder.transform(deep_final_scoring['opponent_team'])
deep_final_scoring[['player_id_encoded', 'season_encoded', 'location_encoded', 'day_slate_encoded']] = ordinal_encoder.transform(deep_final_scoring[['player_id', 'season', 'location', 'day_slate']])
deep_final_scoring[['player_id_encoded', 'season_encoded', 'location_encoded', 'day_slate_encoded']] += 1

# Get number of unique values for each categorical variable
num_player_id = deep_final_scoring['player_id_encoded'].nunique()
print(f'Number of unique player_id: {num_player_id}')
num_game_number = deep_final_scoring['game_number_encoded'].nunique()
print(f'Number of unique game_number: {num_game_number}')
num_team = deep_final_scoring['team_encoded'].nunique()
print(f'Number of unique team: {num_team}')
num_season = deep_final_scoring['season_encoded'].nunique()
print(f'Number of unique season: {num_season}')
num_week = deep_final_scoring['week_encoded'].nunique()
print(f'Number of unique week: {num_week}')
num_opponent_team = deep_final_scoring['opponent_team_encoded'].nunique()
print(f'Number of unique opponent_team: {num_opponent_team}')
num_day_slate = deep_final_scoring['day_slate_encoded'].nunique()
print(f'Number of unique day_slate: {num_day_slate}')
num_location = deep_final_scoring['location_encoded'].nunique()
print(f'Number of unique location: {num_location}')

# Prepare features and target
deep_final_scoring = deep_final_scoring.fillna(0)
deep_final_scoring

Number of unique player_id: 645
Number of unique game_number: 16
Number of unique team: 32
Number of unique season: 1
Number of unique week: 17
Number of unique opponent_team: 32
Number of unique day_slate: 12
Number of unique location: 33

player_id_input_score = deep_final_scoring['player_id_encoded'].values
game_number_input_score = deep_final_scoring['game_number_encoded'].values
team_input_score = deep_final_scoring['team_encoded'].values
season_input_score = deep_final_scoring['season_encoded'].values
week_input_score = deep_final_scoring['week_encoded'].values
opponent_team_input_score = deep_final_scoring['opponent_team_encoded'].values
day_slate_input_score = deep_final_scoring['day_slate_encoded'].values
location_input_score = deep_final_scoring['location_encoded'].values

numeric_features_score = deep_final_scoring[[
    'completions_lagged', 'attempts_lagged', 'passing_yards_lagged',
    'passing_tds_lagged', 'interceptions_lagged', 'sacks_lagged',
    'sack_yards_lagged', 'sack_fumbles_lagged', 'sack_fumbles_lost_lagged',
    'passing_air_yards_lagged', 'passing_yards_after_catch_lagged',
    'passing_first_downs_lagged', 'passing_epa_lagged',
    'passing_2pt_conversions_lagged', 'pacr_lagged', 'dakota_lagged',
    'carries_lagged', 'rushing_yards_lagged', 'rushing_tds_lagged',
    'rushing_fumbles_lagged', 'rushing_fumbles_lost_lagged',
    'rushing_first_downs_lagged', 'rushing_epa_lagged',
    'rushing_2pt_conversions_lagged', 'receptions_lagged', 'targets_lagged',
    'receiving_yards_lagged', 'receiving_tds_lagged',
    'receiving_fumbles_lagged', 'receiving_fumbles_lost_lagged',
    'receiving_air_yards_lagged', 'receiving_yards_after_catch_lagged',
    'receiving_first_downs_lagged', 'receiving_epa_lagged',
    'receiving_2pt_conversions_lagged', 'racr_lagged',
    'target_share_lagged', 'air_yards_share_lagged', 'wopr_lagged',
    'special_teams_tds_lagged', 'fantasy_points_ppr_lagged',
    'used_opponent_avg', 'used_game_number_avg', 'DNP_prior_season_game',
    'never_seen', 'cat_FB', 'cat_QB', 'cat_RB', 'cat_TE', 'cat_WR'
]].values

# Score predictions
predictions_deep25 = deep_model.predict([
    player_id_input_score, game_number_input_score, team_input_score,
    season_input_score, week_input_score, opponent_team_input_score,
    day_slate_input_score, location_input_score, numeric_features_score
])
deep_final_scoring['prediction'] = predictions_deep25.flatten()
deep_final_scoring

379/379 ━━━━━━━━━━━━━━━━━━━━ 1s 2ms/step

# Merging in player_name and grouping by season
deep_results25 = deep_final_scoring.copy()
deep_results25['player_name'] = final_scoring['player_name']
deep_season_predictions = deep_results25.groupby(['player_name', 'season']).sum().reset_index()
deep_season_predictions

/var/folders/7t/jxkc73q943n1l8zb9xmqg9kc0000gn/T/ipykernel_1185/219754331.py:4: FutureWarning:

The default value of numeric_only in DataFrameGroupBy.sum is deprecated. In a future version, numeric_only will default to False. Either specify numeric_only or select only columns which should be valid for the function.

# QBs 2024 Season totals
qbs_2024_season = weekly_stats[(weekly_stats['season'] == 2024) & (weekly_stats['position'] == 'QB')].groupby(['player_display_name', 'season']).sum().reset_index().rename(columns={'player_display_name':'player_name'})

# Top QBs - Neural Network
deep_topQBs = deep_season_predictions[deep_season_predictions['cat_QB'] == 16].sort_values('prediction', ascending=False).head(10)[['player_name', 'prediction']].reset_index(drop=True)
deep_topQBs = deep_topQBs.merge(qbs_2024_season[['player_name', 'fantasy_points_ppr']], how='left', on='player_name').rename(columns={'prediction':'2025 Fantasy Points PPR', 'fantasy_points_ppr':'2024 Fantasy Points PPR'})

# Top QBs - CatBoost
catboost_topQBs = all_25_catboost[all_25_catboost['position'] == 'QB'].groupby(['player_name', 'season']).sum().reset_index().sort_values('prediction', ascending=False).head(10)[['player_name', 'prediction']].reset_index(drop=True)
catboost_topQBs = catboost_topQBs.merge(qbs_2024_season[['player_name', 'fantasy_points_ppr']], how='left', on='player_name').rename(columns={'prediction':'2025 Fantasy Points PPR', 'fantasy_points_ppr':'2024 Fantasy Points PPR'})

# Full comparison df
top_QBs_25 = pd.DataFrame({
    'Rank': np.arange(1, 11),
    'CatBoost QB':catboost_topQBs['player_name'],
    'CatBoost 2025 Fantasy Points':catboost_topQBs['2025 Fantasy Points PPR'],
    'CatBoost QB 2024 Fantasy Points':catboost_topQBs['2024 Fantasy Points PPR'],
    'Neural Network QB':deep_topQBs['player_name'],
    'Neural Network 2025 Fantasy Points':deep_topQBs['2025 Fantasy Points PPR'],
    'Neural Network 2024 Fantasy Points':deep_topQBs['2024 Fantasy Points PPR'],
    '2024 QB':qbs_2024_season.sort_values('fantasy_points_ppr', ascending=False).head(10)['player_name'].values,
    '2024 QB Fantasy Points':qbs_2024_season[['player_name', 'fantasy_points_ppr']].sort_values('fantasy_points_ppr', ascending=False).head(10)['fantasy_points_ppr'].values
})
top_QBs_25.set_index('Rank', inplace=True)
top_QBs_25

/var/folders/7t/jxkc73q943n1l8zb9xmqg9kc0000gn/T/ipykernel_1185/1169808053.py:2: FutureWarning:

The default value of numeric_only in DataFrameGroupBy.sum is deprecated. In a future version, numeric_only will default to False. Either specify numeric_only or select only columns which should be valid for the function.

/var/folders/7t/jxkc73q943n1l8zb9xmqg9kc0000gn/T/ipykernel_1185/1169808053.py:9: FutureWarning:

The default value of numeric_only in DataFrameGroupBy.sum is deprecated. In a future version, numeric_only will default to False. Either specify numeric_only or select only columns which should be valid for the function.

# WRs 2024 Season totals
wrs_2024_season = weekly_stats[(weekly_stats['season'] == 2024) & (weekly_stats['position'] == 'WR')].groupby(['player_display_name', 'season']).sum().reset_index().rename(columns={'player_display_name':'player_name'})

# Top WRs - Neural Network
deep_topWRs = deep_season_predictions[deep_season_predictions['cat_WR'] == 16].sort_values('prediction', ascending=False).head(10)[['player_name', 'prediction']].reset_index(drop=True)
deep_topWRs = deep_topWRs.merge(wrs_2024_season[['player_name', 'fantasy_points_ppr']], how='left', on='player_name').rename(columns={'prediction':'2025 Fantasy Points PPR', 'fantasy_points_ppr':'2024 Fantasy Points PPR'})

# Top WRs - CatBoost
catboost_topWRs = all_25_catboost[all_25_catboost['position'] == 'WR'].groupby(['player_name', 'season']).sum().reset_index().sort_values('prediction', ascending=False).head(10)[['player_name', 'prediction']].reset_index(drop=True)
catboost_topWRs = catboost_topWRs.merge(wrs_2024_season[['player_name', 'fantasy_points_ppr']], how='left', on='player_name').rename(columns={'prediction':'2025 Fantasy Points PPR', 'fantasy_points_ppr':'2024 Fantasy Points PPR'})

# Full comparison df
top_WRs_25 = pd.DataFrame({
    'Rank': np.arange(1, 11),
    'CatBoost WR':catboost_topWRs['player_name'],
    'CatBoost 2025 Fantasy Points':catboost_topWRs['2025 Fantasy Points PPR'],
    'CatBoost WR 2024 Fantasy Points':catboost_topWRs['2024 Fantasy Points PPR'],
    'Neural Network WR':deep_topWRs['player_name'],
    'Neural Network 2025 Fantasy Points':deep_topWRs['2025 Fantasy Points PPR'],
    'Neural Network 2024 Fantasy Points':deep_topWRs['2024 Fantasy Points PPR'],
    '2024 WR':wrs_2024_season.sort_values('fantasy_points_ppr', ascending=False).head(10)['player_name'].values,
    '2024 WR Fantasy Points':wrs_2024_season[['player_name', 'fantasy_points_ppr']].sort_values('fantasy_points_ppr', ascending=False).head(10)['fantasy_points_ppr'].values
})
top_WRs_25.set_index('Rank', inplace=True)
top_WRs_25

/var/folders/7t/jxkc73q943n1l8zb9xmqg9kc0000gn/T/ipykernel_1185/1320571008.py:2: FutureWarning:

The default value of numeric_only in DataFrameGroupBy.sum is deprecated. In a future version, numeric_only will default to False. Either specify numeric_only or select only columns which should be valid for the function.

/var/folders/7t/jxkc73q943n1l8zb9xmqg9kc0000gn/T/ipykernel_1185/1320571008.py:9: FutureWarning:

The default value of numeric_only in DataFrameGroupBy.sum is deprecated. In a future version, numeric_only will default to False. Either specify numeric_only or select only columns which should be valid for the function.

# RBs 2024 Season totals
rbs_2024_season = weekly_stats[(weekly_stats['season'] == 2024) & (weekly_stats['position'].isin(['RB', 'FB']))].groupby(['player_display_name', 'season']).sum().reset_index().rename(columns={'player_display_name':'player_name'})

# Top RBs - Neural Network
deep_topRBs = deep_season_predictions[(deep_season_predictions['cat_RB'] == 16) | (deep_season_predictions['cat_FB'] == 16)].sort_values('prediction', ascending=False).head(10)[['player_name', 'prediction']].reset_index(drop=True)
deep_topRBs = deep_topRBs.merge(rbs_2024_season[['player_name', 'fantasy_points_ppr']], how='left', on='player_name').rename(columns={'prediction':'2025 Fantasy Points PPR', 'fantasy_points_ppr':'2024 Fantasy Points PPR'})

# Top RBs - CatBoost
catboost_topRBs = all_25_catboost[all_25_catboost['position'].isin(['RB', 'FB'])].groupby(['player_name', 'season']).sum().reset_index().sort_values('prediction', ascending=False).head(10)[['player_name', 'prediction']].reset_index(drop=True)
catboost_topRBs = catboost_topRBs.merge(rbs_2024_season[['player_name', 'fantasy_points_ppr']], how='left', on='player_name').rename(columns={'prediction':'2025 Fantasy Points PPR', 'fantasy_points_ppr':'2024 Fantasy Points PPR'})

# Full comparison df
top_RBs_25 = pd.DataFrame({
    'Rank': np.arange(1, 11),
    'CatBoost RB':catboost_topRBs['player_name'],
    'CatBoost 2025 Fantasy Points':catboost_topRBs['2025 Fantasy Points PPR'],
    'CatBoost RB 2024 Fantasy Points':catboost_topRBs['2024 Fantasy Points PPR'],
    'Neural Network RB':deep_topRBs['player_name'],
    'Neural Network 2025 Fantasy Points':deep_topRBs['2025 Fantasy Points PPR'],
    'Neural Network 2024 Fantasy Points':deep_topRBs['2024 Fantasy Points PPR'],
    '2024 RB':rbs_2024_season.sort_values('fantasy_points_ppr', ascending=False).head(10)['player_name'].values,
    '2024 RB Fantasy Points':rbs_2024_season[['player_name', 'fantasy_points_ppr']].sort_values('fantasy_points_ppr', ascending=False).head(10)['fantasy_points_ppr'].values
})
top_RBs_25.set_index('Rank', inplace=True)
top_RBs_25

/var/folders/7t/jxkc73q943n1l8zb9xmqg9kc0000gn/T/ipykernel_1185/2105499564.py:2: FutureWarning:

The default value of numeric_only in DataFrameGroupBy.sum is deprecated. In a future version, numeric_only will default to False. Either specify numeric_only or select only columns which should be valid for the function.

/var/folders/7t/jxkc73q943n1l8zb9xmqg9kc0000gn/T/ipykernel_1185/2105499564.py:9: FutureWarning:

The default value of numeric_only in DataFrameGroupBy.sum is deprecated. In a future version, numeric_only will default to False. Either specify numeric_only or select only columns which should be valid for the function.

# TEs 2024 Season totals
TEs_2024_season = weekly_stats[(weekly_stats['season'] == 2024) & (weekly_stats['position'] == 'TE')].groupby(['player_display_name', 'season']).sum().reset_index().rename(columns={'player_display_name':'player_name'})

# Top TEs - Neural Network
deep_topTEs = deep_season_predictions[deep_season_predictions['cat_TE'] == 16].sort_values('prediction', ascending=False).head(10)[['player_name', 'prediction']].reset_index(drop=True)
deep_topTEs = deep_topTEs.merge(TEs_2024_season[['player_name', 'fantasy_points_ppr']], how='left', on='player_name').rename(columns={'prediction':'2025 Fantasy Points PPR', 'fantasy_points_ppr':'2024 Fantasy Points PPR'})

# Top TEs - CatBoost
catboost_topTEs = all_25_catboost[all_25_catboost['position'] == 'TE'].groupby(['player_name', 'season']).sum().reset_index().sort_values('prediction', ascending=False).head(10)[['player_name', 'prediction']].reset_index(drop=True)
catboost_topTEs = catboost_topTEs.merge(TEs_2024_season[['player_name', 'fantasy_points_ppr']], how='left', on='player_name').rename(columns={'prediction':'2025 Fantasy Points PPR', 'fantasy_points_ppr':'2024 Fantasy Points PPR'})

# Full comparison df
top_TEs_25 = pd.DataFrame({
    'Rank': np.arange(1, 11),
    'CatBoost TE':catboost_topTEs['player_name'],
    'CatBoost 2025 Fantasy Points':catboost_topTEs['2025 Fantasy Points PPR'],
    'CatBoost TE 2024 Fantasy Points':catboost_topTEs['2024 Fantasy Points PPR'],
    'Neural Network TE':deep_topTEs['player_name'],
    'Neural Network 2025 Fantasy Points':deep_topTEs['2025 Fantasy Points PPR'],
    'Neural Network 2024 Fantasy Points':deep_topTEs['2024 Fantasy Points PPR'],
    '2024 TE':TEs_2024_season.sort_values('fantasy_points_ppr', ascending=False).head(10)['player_name'].values,
    '2024 TE Fantasy Points':TEs_2024_season[['player_name', 'fantasy_points_ppr']].sort_values('fantasy_points_ppr', ascending=False).head(10)['fantasy_points_ppr'].values
})
top_TEs_25.set_index('Rank', inplace=True)
top_TEs_25

/var/folders/7t/jxkc73q943n1l8zb9xmqg9kc0000gn/T/ipykernel_1185/2167315060.py:2: FutureWarning:

The default value of numeric_only in DataFrameGroupBy.sum is deprecated. In a future version, numeric_only will default to False. Either specify numeric_only or select only columns which should be valid for the function.

/var/folders/7t/jxkc73q943n1l8zb9xmqg9kc0000gn/T/ipykernel_1185/2167315060.py:9: FutureWarning:

The default value of numeric_only in DataFrameGroupBy.sum is deprecated. In a future version, numeric_only will default to False. Either specify numeric_only or select only columns which should be valid for the function.

%pip install --upgrade google-generativeai

Requirement already satisfied: google-generativeai in ./myenv3.10/lib/python3.10/site-packages (0.8.5)
Requirement already satisfied: tqdm in ./myenv3.10/lib/python3.10/site-packages (from google-generativeai) (4.67.1)
Requirement already satisfied: google-auth>=2.15.0 in ./myenv3.10/lib/python3.10/site-packages (from google-generativeai) (2.40.3)
Requirement already satisfied: protobuf in ./myenv3.10/lib/python3.10/site-packages (from google-generativeai) (4.25.8)
Requirement already satisfied: pydantic in ./myenv3.10/lib/python3.10/site-packages (from google-generativeai) (2.11.5)
Requirement already satisfied: google-api-python-client in ./myenv3.10/lib/python3.10/site-packages (from google-generativeai) (2.171.0)
Requirement already satisfied: typing-extensions in ./myenv3.10/lib/python3.10/site-packages (from google-generativeai) (4.14.0)
Requirement already satisfied: google-api-core in ./myenv3.10/lib/python3.10/site-packages (from google-generativeai) (2.25.0)
Requirement already satisfied: google-ai-generativelanguage==0.6.15 in ./myenv3.10/lib/python3.10/site-packages (from google-generativeai) (0.6.15)
Requirement already satisfied: proto-plus<2.0.0dev,>=1.22.3 in ./myenv3.10/lib/python3.10/site-packages (from google-ai-generativelanguage==0.6.15->google-generativeai) (1.26.1)
Requirement already satisfied: pyasn1-modules>=0.2.1 in ./myenv3.10/lib/python3.10/site-packages (from google-auth>=2.15.0->google-generativeai) (0.4.2)
Requirement already satisfied: cachetools<6.0,>=2.0.0 in ./myenv3.10/lib/python3.10/site-packages (from google-auth>=2.15.0->google-generativeai) (5.5.2)
Requirement already satisfied: rsa<5,>=3.1.4 in ./myenv3.10/lib/python3.10/site-packages (from google-auth>=2.15.0->google-generativeai) (4.9.1)
Requirement already satisfied: requests<3.0.0,>=2.18.0 in ./myenv3.10/lib/python3.10/site-packages (from google-api-core->google-generativeai) (2.32.4)
Requirement already satisfied: googleapis-common-protos<2.0.0,>=1.56.2 in ./myenv3.10/lib/python3.10/site-packages (from google-api-core->google-generativeai) (1.70.0)
Requirement already satisfied: httplib2<1.0.0,>=0.19.0 in ./myenv3.10/lib/python3.10/site-packages (from google-api-python-client->google-generativeai) (0.22.0)
Requirement already satisfied: google-auth-httplib2<1.0.0,>=0.2.0 in ./myenv3.10/lib/python3.10/site-packages (from google-api-python-client->google-generativeai) (0.2.0)
Requirement already satisfied: uritemplate<5,>=3.0.1 in ./myenv3.10/lib/python3.10/site-packages (from google-api-python-client->google-generativeai) (4.2.0)
Requirement already satisfied: annotated-types>=0.6.0 in ./myenv3.10/lib/python3.10/site-packages (from pydantic->google-generativeai) (0.7.0)
Requirement already satisfied: pydantic-core==2.33.2 in ./myenv3.10/lib/python3.10/site-packages (from pydantic->google-generativeai) (2.33.2)
Requirement already satisfied: typing-inspection>=0.4.0 in ./myenv3.10/lib/python3.10/site-packages (from pydantic->google-generativeai) (0.4.1)
Requirement already satisfied: grpcio<2.0.0,>=1.33.2 in ./myenv3.10/lib/python3.10/site-packages (from google-api-core->google-generativeai) (1.73.0)
Requirement already satisfied: grpcio-status<2.0.0,>=1.33.2 in ./myenv3.10/lib/python3.10/site-packages (from google-api-core->google-generativeai) (1.71.0)
Requirement already satisfied: pyparsing!=3.0.0,!=3.0.1,!=3.0.2,!=3.0.3,<4,>=2.4.2 in ./myenv3.10/lib/python3.10/site-packages (from httplib2<1.0.0,>=0.19.0->google-api-python-client->google-generativeai) (3.2.3)
Requirement already satisfied: pyasn1<0.7.0,>=0.6.1 in ./myenv3.10/lib/python3.10/site-packages (from pyasn1-modules>=0.2.1->google-auth>=2.15.0->google-generativeai) (0.6.1)
Requirement already satisfied: charset_normalizer<4,>=2 in ./myenv3.10/lib/python3.10/site-packages (from requests<3.0.0,>=2.18.0->google-api-core->google-generativeai) (3.4.2)
Requirement already satisfied: idna<4,>=2.5 in ./myenv3.10/lib/python3.10/site-packages (from requests<3.0.0,>=2.18.0->google-api-core->google-generativeai) (3.10)
Requirement already satisfied: urllib3<3,>=1.21.1 in ./myenv3.10/lib/python3.10/site-packages (from requests<3.0.0,>=2.18.0->google-api-core->google-generativeai) (2.4.0)
Requirement already satisfied: certifi>=2017.4.17 in ./myenv3.10/lib/python3.10/site-packages (from requests<3.0.0,>=2.18.0->google-api-core->google-generativeai) (2025.4.26)
Collecting protobuf
  Using cached protobuf-5.29.5-cp38-abi3-macosx_10_9_universal2.whl (418 kB)
Installing collected packages: protobuf
  Attempting uninstall: protobuf
    Found existing installation: protobuf 4.25.8
    Uninstalling protobuf-4.25.8:
      Successfully uninstalled protobuf-4.25.8
ERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts.
tensorflow 2.16.2 requires protobuf!=4.21.0,!=4.21.1,!=4.21.2,!=4.21.3,!=4.21.4,!=4.21.5,<5.0.0dev,>=3.20.3, but you have protobuf 5.29.5 which is incompatible.
Successfully installed protobuf-5.29.5

[notice] A new release of pip available: 22.2.2 -> 25.1.1
[notice] To update, run: pip install --upgrade pip
Note: you may need to restart the kernel to use updated packages.

pip install --upgrade google-genai

Requirement already satisfied: google-genai in ./myenv3.10/lib/python3.10/site-packages (1.19.0)
Requirement already satisfied: anyio<5.0.0,>=4.8.0 in ./myenv3.10/lib/python3.10/site-packages (from google-genai) (4.9.0)
Requirement already satisfied: requests<3.0.0,>=2.28.1 in ./myenv3.10/lib/python3.10/site-packages (from google-genai) (2.32.4)
Requirement already satisfied: pydantic<3.0.0,>=2.0.0 in ./myenv3.10/lib/python3.10/site-packages (from google-genai) (2.11.5)
Requirement already satisfied: httpx<1.0.0,>=0.28.1 in ./myenv3.10/lib/python3.10/site-packages (from google-genai) (0.28.1)
Requirement already satisfied: google-auth<3.0.0,>=2.14.1 in ./myenv3.10/lib/python3.10/site-packages (from google-genai) (2.40.3)
Requirement already satisfied: websockets<15.1.0,>=13.0.0 in ./myenv3.10/lib/python3.10/site-packages (from google-genai) (15.0.1)
Requirement already satisfied: typing-extensions<5.0.0,>=4.11.0 in ./myenv3.10/lib/python3.10/site-packages (from google-genai) (4.14.0)
Requirement already satisfied: sniffio>=1.1 in ./myenv3.10/lib/python3.10/site-packages (from anyio<5.0.0,>=4.8.0->google-genai) (1.3.1)
Requirement already satisfied: idna>=2.8 in ./myenv3.10/lib/python3.10/site-packages (from anyio<5.0.0,>=4.8.0->google-genai) (3.10)
Requirement already satisfied: exceptiongroup>=1.0.2 in ./myenv3.10/lib/python3.10/site-packages (from anyio<5.0.0,>=4.8.0->google-genai) (1.3.0)
Requirement already satisfied: rsa<5,>=3.1.4 in ./myenv3.10/lib/python3.10/site-packages (from google-auth<3.0.0,>=2.14.1->google-genai) (4.9.1)
Requirement already satisfied: pyasn1-modules>=0.2.1 in ./myenv3.10/lib/python3.10/site-packages (from google-auth<3.0.0,>=2.14.1->google-genai) (0.4.2)
Requirement already satisfied: cachetools<6.0,>=2.0.0 in ./myenv3.10/lib/python3.10/site-packages (from google-auth<3.0.0,>=2.14.1->google-genai) (5.5.2)
Requirement already satisfied: httpcore==1.* in ./myenv3.10/lib/python3.10/site-packages (from httpx<1.0.0,>=0.28.1->google-genai) (1.0.9)
Requirement already satisfied: certifi in ./myenv3.10/lib/python3.10/site-packages (from httpx<1.0.0,>=0.28.1->google-genai) (2025.4.26)
Requirement already satisfied: h11>=0.16 in ./myenv3.10/lib/python3.10/site-packages (from httpcore==1.*->httpx<1.0.0,>=0.28.1->google-genai) (0.16.0)
Requirement already satisfied: pydantic-core==2.33.2 in ./myenv3.10/lib/python3.10/site-packages (from pydantic<3.0.0,>=2.0.0->google-genai) (2.33.2)
Requirement already satisfied: typing-inspection>=0.4.0 in ./myenv3.10/lib/python3.10/site-packages (from pydantic<3.0.0,>=2.0.0->google-genai) (0.4.1)
Requirement already satisfied: annotated-types>=0.6.0 in ./myenv3.10/lib/python3.10/site-packages (from pydantic<3.0.0,>=2.0.0->google-genai) (0.7.0)
Requirement already satisfied: charset_normalizer<4,>=2 in ./myenv3.10/lib/python3.10/site-packages (from requests<3.0.0,>=2.28.1->google-genai) (3.4.2)
Requirement already satisfied: urllib3<3,>=1.21.1 in ./myenv3.10/lib/python3.10/site-packages (from requests<3.0.0,>=2.28.1->google-genai) (2.4.0)
Requirement already satisfied: pyasn1<0.7.0,>=0.6.1 in ./myenv3.10/lib/python3.10/site-packages (from pyasn1-modules>=0.2.1->google-auth<3.0.0,>=2.14.1->google-genai) (0.6.1)

[notice] A new release of pip available: 22.2.2 -> 25.1.1
[notice] To update, run: pip install --upgrade pip
Note: you may need to restart the kernel to use updated packages.

pip install python-dotenv

Collecting python-dotenv
  Downloading python_dotenv-1.1.0-py3-none-any.whl (20 kB)
Installing collected packages: python-dotenv
Successfully installed python-dotenv-1.1.0

[notice] A new release of pip available: 22.2.2 -> 25.1.1
[notice] To update, run: pip install --upgrade pip
Note: you may need to restart the kernel to use updated packages.

from google import genai
client = genai.Client()

# Iterate through the pager to access individual models
for model in client.models.list():
    print(model.name)

models/embedding-gecko-001
models/gemini-1.0-pro-vision-latest
models/gemini-pro-vision
models/gemini-1.5-pro-latest
models/gemini-1.5-pro-002
models/gemini-1.5-pro
models/gemini-1.5-flash-latest
models/gemini-1.5-flash
models/gemini-1.5-flash-002
models/gemini-1.5-flash-8b
models/gemini-1.5-flash-8b-001
models/gemini-1.5-flash-8b-latest
models/gemini-2.5-pro-exp-03-25
models/gemini-2.5-pro-preview-03-25
models/gemini-2.5-flash-preview-04-17
models/gemini-2.5-flash-preview-05-20
models/gemini-2.5-flash
models/gemini-2.5-flash-preview-04-17-thinking
models/gemini-2.5-flash-lite-preview-06-17
models/gemini-2.5-pro-preview-05-06
models/gemini-2.5-pro-preview-06-05
models/gemini-2.5-pro
models/gemini-2.0-flash-exp
models/gemini-2.0-flash
models/gemini-2.0-flash-001
models/gemini-2.0-flash-exp-image-generation
models/gemini-2.0-flash-lite-001
models/gemini-2.0-flash-lite
models/gemini-2.0-flash-preview-image-generation
models/gemini-2.0-flash-lite-preview-02-05
models/gemini-2.0-flash-lite-preview
models/gemini-2.0-pro-exp
models/gemini-2.0-pro-exp-02-05
models/gemini-exp-1206
models/gemini-2.0-flash-thinking-exp-01-21
models/gemini-2.0-flash-thinking-exp
models/gemini-2.0-flash-thinking-exp-1219
models/gemini-2.5-flash-preview-tts
models/gemini-2.5-pro-preview-tts
models/learnlm-2.0-flash-experimental
models/gemma-3-1b-it
models/gemma-3-4b-it
models/gemma-3-12b-it
models/gemma-3-27b-it
models/gemma-3n-e4b-it
models/embedding-001
models/text-embedding-004
models/gemini-embedding-exp-03-07
models/gemini-embedding-exp
models/aqa
models/imagen-3.0-generate-002
models/veo-2.0-generate-001
models/gemini-2.5-flash-preview-native-audio-dialog
models/gemini-2.5-flash-exp-native-audio-thinking-dialog
models/gemini-2.0-flash-live-001
models/gemini-live-2.5-flash-preview

import os
import pandas as pd
import google.generativeai as genai
from dotenv import load_dotenv

# Load Gemini API key from .env
load_dotenv()
genai.configure(api_key=os.getenv("GEMINI_API_KEY"))
model = genai.GenerativeModel(model_name="gemini-2.0-flash")

def build_context(player_id, season, predictions_df, training_df):
    pred_rows = predictions_df[(predictions_df['player_id'] == player_id) & (predictions_df['season'] == season)]
    train_rows = training_df[training_df['player_id'] == player_id]

    if pred_rows.empty or train_rows.empty:
        return None

    pred_rows = pred_rows.sort_values("week")
    train_rows = train_rows.sort_values("week")

    player_name = pred_rows['player_name'].iloc[0]

    pred_context = pred_rows[
        ['week', 'prediction', 'position', 'team', 'opponent_team', 'game_number',
         'day_slate', 'location', 'used_opponent_avg', 'used_game_number_avg', 
         'DNP_prior_season_game', 'never_seen']
        + [col for col in pred_rows.columns if col.endswith('_lagged')]
    ].to_string(index=False)

    train_context = train_rows[
        ['week', 'fantasy_points_ppr', 'position', 'team', 'opponent_team', 'game_number',
         'day_slate', 'location']
    ].to_string(index=False)

    context = f"""Player: {player_name}, Player ID: {player_id}, Season: {season}

Predictions (weekly):
{pred_context}

Training Data (weekly fantasy performance):
{train_context}

NOTES:
- Columns ending in '_lagged' represent the player's stats from the prior season used as model inputs.
- used_opponent_avg = 1 means prior-season stats were averaged from games against the same opponent.
- used_game_number_avg = 1 means stats were averaged based on same game number.
- DNP_prior_season_game = 1 means the player did not play the equivalent game the previous year.
- never_seen = 1 means the player had no historical data at all from prior seasons.
"""
    return context, player_name

def analyze_prediction(player_id, season, predictions_df, training_df):
    result = build_context(player_id, season, predictions_df, training_df)
    if result is None:
        return "No data available for this player-season.", None

    context, player_name = result

    total_pred = predictions_df[
        (predictions_df['player_id'] == player_id) & (predictions_df['season'] == season)
    ]['prediction'].sum()

    prompt = (
        f"{context}\n\n"
        f"Total predicted fantasy points for this player in {season}: {total_pred:.2f}.\n\n"
        "You are an expert fantasy football analyst writing brief reviews of some predictions on projections of some players' fantasy football points next year. "
        "You don't have understanding of machine learning or deep learning, only masterful understanding of football statistics, player performance, individual player statistics, and "
        "how all of these translate to fantasy football PPR scoring performance. "
        "Evaluate whether the season-long prediction is reasonable given the week-by-week context and the player's season-long totals in the past, "
        "including the features from the prior season used to inform each prediction (all columns ending in '_lagged'). This also means if a player does not have impressive season-long totals in prior seasons, "
        "they should not be expected impressive season-long totals in the future. Likewise with week-to-week performances: if a player does not have impressive stats on a weekly basis in prior seasons,"
        "they should not be expected impressive performances on a week-by-week basis either. Additionally, a player without many stats in the past, you cannot expect them to have great performances in the future. "
        "A player should not be expected a high season-long total if they have not had high season-long totals in the past, regardless of their position and/or team. "
        "Answer in this exact format (no extra line breaks or other formatting changes): "
        "'Rating: [rating out of 100, with 1 being the worst a fantasy prediction can be and 100 being a perfect prediction based on the player's prior season-long performances; "
        "be specific, and give a wide range of ratings (not just in the 65-75 range)]. First reasoning sentence: [reasoning of the rating related to the season-long prediction compared to previous season-long predictions of both the player and players in their position]. "
        "Second reasoning sentence: [reasoning of the rating related to the player's week-by-week performance, calling out certain weeks in which predictions are lower or higher than what might be reasonably "
        "expected given prior performances/lagged stats and how that may contribute to the overall season-long prediction being lower or higher than expected]. Third reasoning sentence: [example or two of any phenomenon or trend noticed "
        "and noted upon in the second reasoning sentence; do not mention any new trends, simply build upon the main trends already touched upon with concrete examples].' (end of example analysis, just the rating and those 3 sentences in that format) "
        "For example, let's say quarterback Matthew Stafford is projected 270 points next season, but has scored at least 300 points in the last several seasons like most quarterbacks of his caliber. "
        "In this next season prediction, he also has a few weeks where he is projected to only score 12 points, which is rather low. Additionally, these lower scoring weeks tend to be against teams "
        "where he has historically performed well, such as teams he has played often (used_opponent_avg = 1) as determined by his fantasy_points_ppr_lagged column against such teams, which is higher than "
        "his average of fantasy_points_ppr_lagged when used_opponent_avg = 0. The output in this case would look like this: "
        "Rating: 58/100. Matthew Stafford is projected to score 270 points, which is on the low end in comparison to the recent seasons where he has scored at least 300 points or more. Certain weeks "
        "he is projected lower point totals than would be expected, such as week 11 where he is projected only 11 points and week 14 where he is projected only 12 points. These lower scoring weeks "
        "are also against teams he has played in the previous season, where he scored 20 and 24 points against them respectively with 300+ yards passing and 3+ touchdowns in each, seeming out of "
        "the ordinary for him and possibly being a source of why his overall season-long prediction is seemingly low. "
    )

    response = model.generate_content(prompt)
    return response.text.strip(), player_name

def generate_analysis(predictions_df, training_df):
    results = []
    for _, row in predictions_df[['player_id', 'season']].drop_duplicates().iterrows():
        explanation, player_name = analyze_prediction(row['player_id'], row['season'], predictions_df, training_df)
        results.append({
            'player_id': row['player_id'],
            'player_name': player_name,
            'season': row['season'],
            'gemini_analysis': explanation
        })
    return pd.DataFrame(results)

catboost_top_all = pd.concat([catboost_topQBs, catboost_topRBs, catboost_topWRs, catboost_topTEs])
catboost_top_all.reset_index(drop=True, inplace=True)
catboost_top_all

analysis_catboost_qbs = generate_analysis(all_25_catboost[all_25_catboost['player_name'].isin(catboost_topQBs['player_name'].values)], 
                                      lagged_stats[lagged_stats['player_name'].isin(catboost_topQBs['player_name'].values)])
analysis_catboost_qbs

analysis_catboost_wrs = generate_analysis(all_25_catboost[all_25_catboost['player_name'].isin(catboost_topWRs['player_name'].values)], 
                                      lagged_stats[lagged_stats['player_name'].isin(catboost_topWRs['player_name'].values)])
analysis_catboost_wrs

analysis_catboost_RBs = generate_analysis(all_25_catboost[all_25_catboost['player_name'].isin(catboost_topRBs['player_name'].values)], 
                                      lagged_stats[lagged_stats['player_name'].isin(catboost_topRBs['player_name'].values)])
analysis_catboost_RBs

analysis_catboost_TEs = generate_analysis(all_25_catboost[all_25_catboost['player_name'].isin(catboost_topTEs['player_name'].values)], 
                                      lagged_stats[lagged_stats['player_name'].isin(catboost_topTEs['player_name'].values)])
analysis_catboost_TEs

all_25_deep = all_25_catboost.drop(columns=['prediction'])
all_25_deep['prediction'] = deep_results25['prediction']
all_25_deep

deep_top_all = pd.concat([deep_topQBs, deep_topRBs, deep_topWRs, deep_topTEs])
deep_top_all.reset_index(drop=True, inplace=True)
deep_top_all

analysis_deep_qbs = generate_analysis(all_25_deep[all_25_deep['player_name'].isin(deep_topQBs['player_name'].values)], 
                                      lagged_stats[lagged_stats['player_name'].isin(deep_topQBs['player_name'].values)])
analysis_deep_qbs

analysis_deep_WRs = generate_analysis(all_25_deep[all_25_deep['player_name'].isin(deep_topWRs['player_name'].values)], 
                                      lagged_stats[lagged_stats['player_name'].isin(deep_topWRs['player_name'].values)])
analysis_deep_WRs

analysis_deep_RBs = generate_analysis(all_25_deep[all_25_deep['player_name'].isin(deep_topRBs['player_name'].values)], 
                                      lagged_stats[lagged_stats['player_name'].isin(deep_topRBs['player_name'].values)])
analysis_deep_RBs

analysis_deep_TEs = generate_analysis(all_25_deep[all_25_deep['player_name'].isin(deep_topTEs['player_name'].values)], 
                                      lagged_stats[lagged_stats['player_name'].isin(deep_topTEs['player_name'].values)])
analysis_deep_TEs

# Combining all analyzed Top 10s
analysis_catboost_all = pd.concat([analysis_catboost_qbs, analysis_catboost_wrs, analysis_catboost_RBs, analysis_catboost_TEs])

# Adding Rank and Position columns to MLP Top 10s
positions = ['QB'] * 10 + ['RB'] * 10 + ['WR'] * 10 + ['TE'] * 10
catboost_top_all['Position'] = positions

# Compute Position Rank (1 = highest points within that position)
catboost_top_all['Position Rank'] = catboost_top_all.groupby('Position')['2025 Fantasy Points PPR'].rank(method='first', ascending=False).astype(int)
catboost_top_all_analyzed = catboost_top_all.merge(analysis_catboost_all[['player_name', 'gemini_analysis']], how='left', on='player_name')
catboost_top_all_analyzed

!pip install tabulate

Collecting tabulate
  Downloading tabulate-0.9.0-py3-none-any.whl (35 kB)
Installing collected packages: tabulate
Successfully installed tabulate-0.9.0

[notice] A new release of pip available: 22.2.2 -> 25.1.1
[notice] To update, run: pip install --upgrade pip

# Saving off as markdown file
with open("CatBoost Analyzed Results.md", "w") as f:
    f.write(catboost_top_all_analyzed.to_markdown(index=False))

# Combining all analyzed Top 10s
analysis_deep_all = pd.concat([analysis_deep_qbs, analysis_deep_WRs, analysis_deep_RBs, analysis_deep_TEs])

# Adding Rank and Position columns to MLP Top 10s
positions = ['QB'] * 10 + ['RB'] * 10 + ['WR'] * 10 + ['TE'] * 10
deep_top_all['Position'] = positions

# Compute Position Rank (1 = highest points within that position)
deep_top_all['Position Rank'] = deep_top_all.groupby('Position')['2025 Fantasy Points PPR'].rank(method='first', ascending=False).astype(int)
deep_top_all_analyzed = deep_top_all.merge(analysis_deep_all[['player_name', 'gemini_analysis']], how='left', on='player_name')
deep_top_all_analyzed

# Saving off as markdown file
with open("Neural Network Analyzed Results.md", "w") as f:
    f.write(deep_top_all_analyzed.to_markdown(index=False))

import nfl_data_py as nfl
import pandas as pd
import numpy as np
weekly_stats_plus = nfl.import_weekly_data(years = range(2010, 2025))
weekly_stats_plus.rename(columns={'recent_team':'team'}, inplace=True)
weekly_stats_plus

Downcasting floats.

schedules = nfl.import_schedules(years = range(2010, 2025))
schedules['time'] = pd.to_datetime(schedules['gametime'])
schedules['hour'] = schedules['time'].dt.hour
schedules['slate'] = np.where(
    schedules['hour'].isin([12, 13, 14]),
    'Morning',
    np.where(
        schedules['hour'].isin([15]),
        'Noon',
        np.where(
            schedules['hour'].isin([16, 17, 18, 19]),
            'Afternoon',
            np.where(
                schedules['hour'].isin([20, 21, 22, 23]),
                'Night',
                'global'
            )
        )
    )
)

# Unifying abbreviations of teams that have moved (and thus changed their abbrevaitions)
team_map = {'STL':'LA', 'SD':'LAC', 'OAK':'LV'}
schedules['home_team'] = schedules['home_team'].replace(team_map)
schedules['away_team'] = schedules['away_team'].replace(team_map)

schedules['day_slate'] = schedules['weekday'] + '_' + schedules['slate']

schedules['location'] = np.where(
    schedules['stadium'].isin(['Wembley Stadium', 'Rogers Centre', 'Twickenham Stadium', 'Azteca Stadium', 'Tottenham Stadium',
                                  'Allianz Arena', 'Deutsche Bank Park', 'Arena Corinthians']), 'global', schedules['home_team'])

schedules

weekly_stats_plus['matchup'] = weekly_stats_plus.apply(lambda row: '_'.join(sorted([row['team'], row['opponent_team']])), axis=1)
schedules['matchup'] = schedules.apply(lambda row: '_'.join(sorted([row['home_team'], row['away_team']])), axis=1)
weekly_stats_plus = weekly_stats_plus.merge(schedules[['season', 'week', 'matchup', 'game_type', 'day_slate', 'location', 
                                             'div_game', 'home_coach', 'away_coach', 'home_team']], how='left', on=['season', 'week', 'matchup'])

weekly_stats_plus['at_home'] = np.where(weekly_stats_plus['team'] == weekly_stats_plus['location'], 1, 0)
weekly_stats_plus['coach'] = np.where(weekly_stats_plus['home_team'] == weekly_stats_plus['team'], weekly_stats_plus['home_coach'], weekly_stats_plus['away_coach'])
weekly_stats_plus['opponent_coach'] = np.where(weekly_stats_plus['home_team'] == weekly_stats_plus['team'], weekly_stats_plus['away_coach'], weekly_stats_plus['home_coach'])
weekly_stats_plus.drop(columns = ['home_coach', 'away_coach', 'home_team'], inplace=True)
weekly_stats_plus

game_no = weekly_stats_plus[['season', 'week', 'team', 'location']].groupby(['season', 'team', 'week']).agg({'location': 'first'})
game_no['game_number'] = game_no.groupby(['season', 'team']).cumcount() + 1
game_no.reset_index(inplace=True)
weekly_stats_plus = weekly_stats_plus.merge(game_no, how='left', on=['season', 'team', 'week', 'location'])
weekly_stats_plus

# Reformatting rows and columns
weekly_stats_plus = weekly_stats_plus.sort_values(['season', 'week'])
cols = list(weekly_stats_plus.columns)
weekly_stats_plus = weekly_stats_plus[cols[:3] + [cols[-1]] + cols[3:-1]]
weekly_stats_plus.reset_index(inplace=True, drop=True)
weekly_stats_plus

# Limiting to fantasy weeks only
weekly_stats_plus = weekly_stats_plus[
    ((weekly_stats_plus['season'] <= 2020) & (weekly_stats_plus['week'] <= 16)) | 
    ((weekly_stats_plus['season'] >= 2021) & (weekly_stats_plus['week'] <= 17))
]

# Limiting to relevant positions only
weekly_stats_plus = weekly_stats_plus[weekly_stats_plus['position_group'].isin(['WR', 'QB', 'TE', 'FB', 'RB'])]

# Limiting to relevant rows only
weekly_stats_plus = weekly_stats_plus.drop(columns={
    'position_group', 'headshot_url', 'player_name', 'season_type', 'matchup', 'game_type'
})
weekly_stats_plus = weekly_stats_plus[~weekly_stats_plus['game_number'].isna()]
weekly_stats_plus.reset_index(inplace=True, drop=True)
weekly_stats_plus

# Define meta/stat columns
meta_cols = ['player_id', 'player_display_name', 'position', 'team', 'season', 'week', 'game_number', 'opponent_team']
stat_cols = [col for col in weekly_stats_plus.columns if col not in meta_cols + ['day_slate', 'location', 'coach', 'opponent_coach', 'div_game', 'at_home']]

# Copy original dataframe
current_df_plus = weekly_stats_plus.copy()
prior_df_plus = weekly_stats_plus.copy()
prior_df_plus['season'] += 1  # Offset to match with current season

# Step 1: Get average stats vs each opponent in the prior season
opponent_avg_plus = (
    prior_df_plus
    .groupby(['player_id', 'season', 'opponent_team'])[stat_cols]
    .mean()
    .reset_index()
    .rename(columns={col: f"{col}_lag_opponent" for col in stat_cols})
)

# Step 2: Merge opponent-based lag stats into current_df_plus
merged_plus = current_df_plus.merge(
    opponent_avg_plus,
    how='left',
    left_on=['player_id', 'season', 'opponent_team'],
    right_on=['player_id', 'season', 'opponent_team']
)

# Step 3: Identify rows with missing opponent lag values
missing_opponent = merged_plus[[f"{col}_lag_opponent" for col in stat_cols]].isna().all(axis=1)

# Step 4: Prepare game_number-based fallback from prior_df_plus using smart moving average
expanded_frames = []

for g in range(1, 17):
    if g == 1:
        games_to_use = [1, 2, 3]
    elif g == 16:
        games_to_use = [14, 15, 16]
    else:
        games_to_use = [g - 1, g, g + 1]

    temp = (
        prior_df_plus[prior_df_plus['game_number'].isin(games_to_use)]
        .groupby(['player_id', 'season'])[stat_cols]
        .mean()
        .reset_index()
    )
    temp['game_number'] = g  # Assign the center game_number
    expanded_frames.append(temp)

fallback_avg = pd.concat(expanded_frames, ignore_index=True)
fallback_avg = fallback_avg.rename(columns={col: f"{col}_lag_game" for col in stat_cols})

# Step 5: Merge fallback lag stats only to missing rows
fallback_df = current_df_plus.loc[missing_opponent, ['player_id', 'season', 'game_number']]
fallback_df = fallback_df.merge(
    fallback_avg,
    how='left',
    on=['player_id', 'season', 'game_number']
)

# Step 6: Merge fallback_df back into merged_plus DataFrame
for col in stat_cols:
    lag_col = f"{col}_lag_opponent"
    fallback_col = f"{col}_lag_game"
    merged_plus.loc[missing_opponent, lag_col] = fallback_df[fallback_col].values

# Step 7: Add integer indicators for the source of the lag
merged_plus['used_opponent_avg_plus'] = (~missing_opponent).astype(int)
merged_plus['used_game_number_avg'] = missing_opponent.astype(int)

# Step 8: Identify rows with no lagged data (DNPs)
merged_plus['DNP_prior_season_game'] = merged_plus[f"{stat_cols[0]}_lag_opponent"].isna().astype(int)

# Step 9: Fill with player prior season averages where DNP occurred
prior_season_avg_plus = (
    weekly_stats_plus
    .groupby(['player_id', 'season'])[stat_cols]
    .mean()
    .reset_index()
    .rename(columns={col: f"{col}_lag_avg_season" for col in stat_cols})
)

season_avg_df_plus = current_df_plus.loc[merged_plus['DNP_prior_season_game'] == 1, ['player_id', 'season']].copy()
season_avg_df_plus['season'] -= 1  # Adjust to prior season
season_avg_df_plus = season_avg_df_plus.merge(
    prior_season_avg_plus,
    how='left',
    on=['player_id', 'season']
)

# Fill DNP rows with prior season averages
for col in stat_cols:
    merged_plus.loc[merged_plus['DNP_prior_season_game'] == 1, f"{col}_lag_opponent"] = season_avg_df_plus[f"{col}_lag_avg_season"].values

# Step 10: Identify "never_seen" players — no prior season record
seen_players_plus = weekly_stats_plus[['player_id', 'season']].drop_duplicates()
seen_players_plus['season'] += 1
merged_plus = merged_plus.merge(
    seen_players_plus.assign(seen_before=1),
    on=['player_id', 'season'],
    how='left'
)
merged_plus['never_seen'] = merged_plus['seen_before'].isna().astype(int)
merged_plus.drop(columns='seen_before', inplace=True)

# Final lagged dataframe
lagged_stats_plus = merged_plus.copy()

# Deleting extra columns
cols_to_drop = list(lagged_stats_plus.loc[:, 'completions':'fantasy_points'].columns) + ['fantasy_points_lag_opponent']
lagged_stats_plus = lagged_stats_plus.drop(columns=cols_to_drop)
lagged_stats_plus.rename(columns=lambda col: col.replace('lag_opponent', 'lagged'), inplace=True)
lagged_stats_plus.rename(columns={'player_display_name': 'player_name'}, inplace=True)
lagged_stats_plus.loc[lagged_stats_plus['position'] == 'HB', 'position'] = 'RB'

# Getting rid of null game_number columns - wrongly input data
lagged_stats_plus = lagged_stats_plus[lagged_stats_plus['game_number'].notna()]
lagged_stats_plus['season'] = lagged_stats_plus['season'].astype(str)
lagged_stats_plus['week'] = lagged_stats_plus['week'].astype(str)
lagged_stats_plus['game_number'] = lagged_stats_plus['game_number'].astype(str)
lagged_stats_plus['yards_per_carry_lagged'] = lagged_stats_plus['rushing_yards_lagged'] / lagged_stats_plus['carries_lagged']
lagged_stats_plus['yards_per_attempt_lagged'] = lagged_stats_plus['passing_yards_lagged'] / lagged_stats_plus['attempts_lagged']
lagged_stats_plus['completion_percentage_lagged'] = lagged_stats_plus['completions_lagged'] / lagged_stats_plus['attempts_lagged']
lagged_stats_plus['yards_per_reception_lagged'] = lagged_stats_plus['receiving_yards_lagged'] / lagged_stats_plus['receptions_lagged']
lagged_stats_plus

all_catboost_df_plus = lagged_stats_plus.copy()
all_catboost_df_plus['season'] = all_catboost_df_plus['season'].astype(str)
all_catboost_df_plus['week'] = all_catboost_df_plus['week'].astype(str)
all_catboost_df_plus['game_number'] = all_catboost_df_plus['game_number'].astype(str)

# Training split
all_train_plus = all_catboost_df_plus.sample(n=round(len(all_catboost_df_plus)*.7), random_state=99)
all_train_plus_X = all_train_plus.loc[:, all_train_plus.columns!='fantasy_points_ppr']
all_train_plus_y = all_train_plus['fantasy_points_ppr']
all_train_plus_indices = all_train_plus.index

all_nontrain_plus = all_catboost_df_plus.drop(all_train_plus.index)

# test split
all_test_plus = all_nontrain_plus.sample(n=round(len(all_nontrain_plus)*.5), random_state=99)
all_test_plus_X = all_test_plus.loc[:, all_test_plus.columns!='fantasy_points_ppr']
all_test_plus_y = all_test_plus['fantasy_points_ppr']
all_test_plus_X_indices = all_test_plus_X.index

# Validation split
all_val_plus = all_nontrain_plus.drop(all_test_plus.index)
all_val_plus_X = all_val_plus.loc[:, all_val_plus.columns!='fantasy_points_ppr']
all_val_plus_y = all_val_plus['fantasy_points_ppr']
all_val_plus

# Parameter grid
iterations_ls = [500, 1000]
learning_rates_ls = [0.01, 0.03, 0.1]
depth_list = [4,6,8]

all_cv_results_plus = []

all_cvpool_plus = Pool(data=all_train_plus_X, label=all_train_plus_y, cat_features=['player_id', 'player_name', 'position', 'team', 'opponent_team', 'day_slate', 'location', 'coach', 'opponent_coach'])

# Grid search
for i, l, d in tqdm(product(iterations_ls, learning_rates_ls, depth_list),
                    total = len(iterations_ls)*len(learning_rates_ls)*len(depth_list)):
        all_params = {'loss_function':'MAE',
                     'iterations': i,          
                     'learning_rate': l,       
                     'depth': d,
                     'early_stopping_rounds': 50,
                     'verbose': False, }                  
        all_cvscores_plus = cv(pool=all_cvpool_plus, params = all_params, fold_count = 5, shuffle=True, logging_level='Silent', partition_random_seed=30)
        best_all_cvscore_plus = all_cvscores_plus['test-MAE-mean'].min()
        all_cv_results_plus.append({'iterations': i,
                              'learning_rate': l,
                              'depth': d,
                              'mae': best_all_cvscore_plus})

all_cv_results_plus_df = pd.DataFrame(all_cv_results_plus)
best_params_all_plus = all_cv_results_plus_df.sort_values(by='mae').iloc[0]
best_params_all_plus

100%|██████████| 18/18 [31:01<00:00, 103.42s/it]

iterations       1000.000000
learning_rate       0.030000
depth               8.000000
mae                 4.760972
Name: 14, dtype: float64

# Training Model on Full Training Set
final_all_params_plus = {
    'loss_function':'MAE',
    'iterations': int(best_params_all_plus['iterations']),
    'learning_rate': float(best_params_all_plus['learning_rate']),
    'depth': int(best_params_all_plus['depth']),
    'early_stopping_rounds': 50,
    'verbose': False
}

all_catboost_plus = CatBoostRegressor(**final_all_params_plus)
all_catboost_plus.fit(all_train_plus_X, all_train_plus_y, cat_features=['player_id', 'player_name', 'position', 'team', 'opponent_team', 'day_slate', 'location', 'coach', 'opponent_coach'], 
                eval_set=(all_val_plus_X, all_val_plus_y))

<catboost.core.CatBoostRegressor at 0x1528727d0>

all_val_plus_preds = all_catboost_plus.predict(all_val_plus_X)
print(f'Validation MAE: {mean_absolute_error(all_val_plus_y, all_val_plus_preds)}')
print(f'Validation R2: {r2_score(all_val_plus_y, all_val_plus_preds)}')

Validation MAE: 4.722122080630746
Validation R2: 0.32052363711854226

all_test_plus_preds = all_catboost_plus.predict(all_test_plus_X)
print(f'test MAE: {mean_absolute_error(all_test_plus_y, all_test_plus_preds)}')
print(f'test R2: {r2_score(all_test_plus_y, all_test_plus_preds)}')

# Predictions with indices
all_test_plus_preds_wind = np.column_stack((all_test_plus_X_indices, all_test_plus_preds))
all_sorted_plus_preds = all_test_plus_preds_wind[all_test_plus_preds_wind[:,0].argsort()]
all_sorted_plus_preds_df = pd.DataFrame(all_sorted_plus_preds, columns = ['index', 'prediction'])
all_sorted_plus_preds_df.set_index('index', inplace=True)
all_sorted_plus_preds_v_test = pd.concat([all_test_plus_X, all_sorted_plus_preds_df], axis = 1)
all_test_plus_preds_exp = pd.concat([all_sorted_plus_preds_v_test, all_test_plus_y], axis=1)
all_test_plus_preds_exp

test MAE: 4.7841822458012055
test R2: 0.3278022112330933

explainer_all_plus = shap.TreeExplainer(all_catboost_plus, feature_perturbation='tree_path_dependent')
shap_values_all_plus = explainer_all_plus(all_test_plus_X)
shap_values_all_plus

.values =
array([[-1.94114044e+00, -1.69303749e-01, -6.26873978e-03, ...,
        -1.00473085e-02, -7.45635472e-03,  1.72369931e-03],
       [-1.28410990e+00, -4.87931659e-01,  2.48163303e-03, ...,
        -5.78751517e-03, -3.15676670e-03, -1.58703186e-02],
       [ 3.09549826e+00,  1.13646900e+00, -1.30160529e-02, ...,
        -4.42463713e-03,  1.05252488e-01,  1.70253653e-02],
       ...,
       [-2.07236097e+00, -4.09575484e-01, -1.82245356e-02, ...,
        -7.98084634e-03,  1.76266134e-04, -5.04219365e-03],
       [ 3.70335992e+00,  9.84480948e-01,  2.47414301e-02, ...,
        -1.69223290e-02, -3.80250658e-03, -1.66831571e-02],
       [-1.85151298e+00, -3.52464883e-01, -1.42652408e-02, ...,
        -9.49013003e-03, -8.77592404e-03, -4.60739284e-03]])

.base_values =
array([7.39647813, 7.39647813, 7.39647813, ..., 7.39647813, 7.39647813,
       7.39647813])

.data =
array([['00-0029141', 'Alfred Morris', '4.0', ..., nan, nan, 13.0],
       ['00-0025463', 'Jason Hill', '10.0', ..., nan, nan, 15.0],
       ['00-0036212', 'Tua Tagovailoa', '2.0', ..., 8.057142857142857,
        0.7142857142857143, nan],
       ...,
       ['00-0026286', 'Jerome Felton', '1.0', ..., nan, nan, nan],
       ['00-0030496', "Le'Veon Bell", '6.0', ..., nan, nan,
        4.0000001192092896],
       ['00-0034279', 'Equanimeous St. Brown', '3.0', ..., nan, nan, 0.0]],
      dtype=object)

shap.plots.beeswarm(shap_values_all_plus, max_display = 25)

	game_id	season	game_type	week	gameday	weekday	gametime	away_team	away_score	home_team	home_score	location	result	total	overtime	old_game_id	gsis	nfl_detail_id	pfr	pff	espn	ftn	away_rest	home_rest	away_moneyline	home_moneyline	spread_line	away_spread_odds	home_spread_odds	total_line	under_odds	over_odds	div_game	roof	surface	temp	wind	away_qb_id	home_qb_id	away_qb_name	home_qb_name	away_coach	home_coach	referee	stadium_id	stadium	time	hour	slate	day_slate
2913	2010_01_MIN_NO	2010	REG	1	2010-09-09	Thursday	20:30	MIN	9.0	NO	14.0	NO	5.0	23.0	0.0	2010090900	54863.0	NaN	201009090nor	1727.0	300909018.0	NaN	7	7	197.0	-220.0	4.5	-105.0	-103.0	48.5	-104.0	-106.0	0	dome	sportturf	NaN	NaN	00-0005106	00-0020531	Brett Favre	Drew Brees	Brad Childress	Sean Payton	Terry McAulay	NOR00	Louisiana Superdome	2025-06-24 20:30:00	20	Night	Thursday_Night
2914	2010_01_MIA_BUF	2010	REG	1	2010-09-12	Sunday	13:00	MIA	15.0	BUF	10.0	BUF	-5.0	25.0	0.0	2010091201	54864.0	NaN	201009120buf	1729.0	300912002.0	NaN	7	7	-155.0	140.0	-3.0	-106.0	-102.0	39.5	-110.0	100.0	1	outdoors	astroplay	62.0	7.0	00-0026197	00-0025479	Chad Henne	Trent Edwards	Tony Sparano	Chan Gailey	Clete Blakeman	BUF00	Ralph Wilson Stadium	2025-06-24 13:00:00	13	Morning	Sunday_Morning
2915	2010_01_DET_CHI	2010	REG	1	2010-09-12	Sunday	13:00	DET	14.0	CHI	19.0	CHI	5.0	33.0	0.0	2010091207	54865.0	NaN	201009120chi	1736.0	300912003.0	NaN	7	7	248.0	-280.0	6.5	103.0	-111.0	44.5	-105.0	-105.0	1	outdoors	grass	75.0	10.0	00-0026498	00-0024226	Matthew Stafford	Jay Cutler	Jim Schwartz	Lovie Smith	Gene Steratore	CHI98	Soldier Field	2025-06-24 13:00:00	13	Morning	Sunday_Morning
2916	2010_01_IND_HOU	2010	REG	1	2010-09-12	Sunday	13:00	IND	24.0	HOU	34.0	HOU	10.0	58.0	0.0	2010091203	54866.0	NaN	201009120htx	1731.0	300912034.0	NaN	7	7	-117.0	106.0	-1.0	-110.0	102.0	47.5	-102.0	-108.0	1	closed	grass	NaN	NaN	00-0010346	00-0022787	Peyton Manning	Matt Schaub	Jim Caldwell	Gary Kubiak	Ed Hochuli	HOU00	Reliant Stadium	2025-06-24 13:00:00	13	Morning	Sunday_Morning
2917	2010_01_DEN_JAX	2010	REG	1	2010-09-12	Sunday	13:00	DEN	17.0	JAX	24.0	JAX	7.0	41.0	0.0	2010091204	54867.0	NaN	201009120jax	1732.0	300912030.0	NaN	7	7	166.0	-185.0	3.0	109.0	-118.0	41.5	-110.0	100.0	0	outdoors	grass	90.0	10.0	00-0023541	00-0021231	Kyle Orton	David Garrard	Josh McDaniels	Jack Del Rio	Walt Coleman	JAX00	EverBank Field	2025-06-24 13:00:00	13	Morning	Sunday_Morning
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
6986	2024_20_LA_PHI	2024	DIV	20	2025-01-19	Sunday	15:00	LA	22.0	PHI	28.0	PHI	6.0	50.0	0.0	2025011900	59837.0	NaN	202501190phi	NaN	401671937.0	NaN	6	7	260.0	-325.0	7.0	-112.0	-108.0	43.5	-110.0	-110.0	0	outdoors	grass	34.0	8.0	00-0026498	00-0036389	Matthew Stafford	Jalen Hurts	Sean McVay	Nick Sirianni	Shawn Smith	PHI00	Lincoln Financial Field	2025-06-24 15:00:00	15	Noon	Sunday_Noon
6987	2024_20_BAL_BUF	2024	DIV	20	2025-01-19	Sunday	18:30	BAL	25.0	BUF	27.0	BUF	2.0	52.0	0.0	2025011901	59838.0	NaN	202501190buf	NaN	401671886.0	NaN	8	7	-118.0	-102.0	-1.5	-102.0	-118.0	51.5	-105.0	-115.0	0	outdoors	a_turf	19.0	4.0	00-0034796	00-0034857	Lamar Jackson	Josh Allen	John Harbaugh	Sean McDermott	Carl Cheffers	BUF00	New Era Field	2025-06-24 18:30:00	18	Afternoon	Sunday_Afternoon
6988	2024_21_WAS_PHI	2024	CON	21	2025-01-26	Sunday	15:00	WAS	23.0	PHI	55.0	PHI	32.0	78.0	0.0	2025012600	59839.0	NaN	202501260phi	NaN	401671888.0	NaN	8	7	205.0	-250.0	6.0	-110.0	-110.0	46.5	-105.0	-115.0	1	outdoors	grass	40.0	11.0	00-0039910	00-0036389	Jayden Daniels	Jalen Hurts	Dan Quinn	Nick Sirianni	Shawn Hochuli	PHI00	Lincoln Financial Field	2025-06-24 15:00:00	15	Noon	Sunday_Noon
6989	2024_21_BUF_KC	2024	CON	21	2025-01-26	Sunday	18:30	BUF	29.0	KC	32.0	KC	3.0	61.0	0.0	2025012601	59840.0	NaN	202501260kan	NaN	401671887.0	NaN	7	8	-102.0	-118.0	1.5	-115.0	-105.0	49.5	-115.0	-105.0	0	outdoors	grass	31.0	8.0	00-0034857	00-0033873	Josh Allen	Patrick Mahomes	Sean McDermott	Andy Reid	Clete Blakeman	KAN00	GEHA Field at Arrowhead Stadium	2025-06-24 18:30:00	18	Afternoon	Sunday_Afternoon
6990	2024_22_KC_PHI	2024	SB	22	2025-02-09	Sunday	18:30	KC	22.0	PHI	40.0	PHI	18.0	62.0	0.0	2025020900	59841.0	NaN	202502090phi	NaN	401671889.0	NaN	14	14	-120.0	100.0	-1.5	-105.0	-115.0	48.5	-105.0	-115.0	0	dome	sportturf	NaN	NaN	00-0033873	00-0036389	Patrick Mahomes	Jalen Hurts	Andy Reid	Nick Sirianni	Ron Torbert	NOR00	Mercedes-Benz Superdome	2025-06-24 18:30:00	18	Afternoon	Sunday_Afternoon

	player_id	player_name	game_number	position	team	season	week	opponent_team	fantasy_points_ppr	day_slate	location	completions_lagged	attempts_lagged	passing_yards_lagged	passing_tds_lagged	interceptions_lagged	sacks_lagged	sack_yards_lagged	sack_fumbles_lagged	sack_fumbles_lost_lagged	passing_air_yards_lagged	passing_yards_after_catch_lagged	passing_first_downs_lagged	passing_epa_lagged	passing_2pt_conversions_lagged	pacr_lagged	dakota_lagged	carries_lagged	rushing_yards_lagged	rushing_tds_lagged	rushing_fumbles_lagged	rushing_fumbles_lost_lagged	rushing_first_downs_lagged	rushing_epa_lagged	rushing_2pt_conversions_lagged	receptions_lagged	targets_lagged	receiving_yards_lagged	receiving_tds_lagged	receiving_fumbles_lagged	receiving_fumbles_lost_lagged	receiving_air_yards_lagged	receiving_yards_after_catch_lagged	receiving_first_downs_lagged	receiving_epa_lagged	receiving_2pt_conversions_lagged	racr_lagged	target_share_lagged	air_yards_share_lagged	wopr_lagged	special_teams_tds_lagged	fantasy_points_ppr_lagged	used_opponent_avg	used_game_number_avg	DNP_prior_season_game	never_seen
0	00-0004161	Jake Delhomme	1.0	QB	CLE	2010	1	TB	9.180000	Sunday_Morning	TB	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	0	1	1	1
1	00-0004541	Donald Driver	1.0	WR	GB	2010	1	PHI	14.000000	Sunday_Afternoon	PHI	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	0	1	1	1
2	00-0005091	Kevin Faulk	1.0	RB	NE	2010	1	CIN	11.000000	Sunday_Morning	NE	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	0	1	1	1
3	00-0005106	Brett Favre	1.0	QB	MIN	2010	1	NO	8.840000	Thursday_Night	NO	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	0	1	1	1
4	00-0005231	Brian Finneran	1.0	WR	ATL	2010	1	PIT	0.000000	Sunday_Morning	PIT	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	0	1	1	1
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
71975	00-0039915	Ladd McConkey	16.0	WR	LAC	2024	17	NE	29.400000	Saturday_Morning	NE	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	0	1	1	1
71976	00-0039916	Ricky Pearsall	16.0	WR	SF	2024	17	DET	28.700001	Monday_Night	SF	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	0	1	1	1
71977	00-0039917	Michael Penix	16.0	QB	ATL	2024	17	WAS	11.220000	Sunday_Night	WAS	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	0	1	1	1
71978	00-0039918	Caleb Williams	16.0	QB	CHI	2024	17	SEA	6.580000	Thursday_Night	CHI	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	0	1	1	1
71979	00-0039919	Rome Odunze	16.0	WR	CHI	2024	17	SEA	2.500000	Thursday_Night	CHI	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	0	1	1	1

	fantasy_points_ppr	completions_lagged	attempts_lagged	passing_yards_lagged	passing_tds_lagged	interceptions_lagged	sacks_lagged	sack_yards_lagged	sack_fumbles_lagged	sack_fumbles_lost_lagged	passing_air_yards_lagged	passing_yards_after_catch_lagged	passing_first_downs_lagged	passing_epa_lagged	passing_2pt_conversions_lagged	pacr_lagged	dakota_lagged	carries_lagged	rushing_yards_lagged	rushing_tds_lagged	rushing_fumbles_lagged	rushing_fumbles_lost_lagged	rushing_first_downs_lagged	rushing_epa_lagged	rushing_2pt_conversions_lagged	receptions_lagged	targets_lagged	receiving_yards_lagged	receiving_tds_lagged	receiving_fumbles_lagged	receiving_fumbles_lost_lagged	receiving_air_yards_lagged	receiving_yards_after_catch_lagged	receiving_first_downs_lagged	receiving_epa_lagged	receiving_2pt_conversions_lagged	racr_lagged	target_share_lagged	air_yards_share_lagged	wopr_lagged	special_teams_tds_lagged	fantasy_points_ppr_lagged	used_opponent_avg	used_game_number_avg	DNP_prior_season_game	never_seen
count	71934.000000	53894.000000	53894.000000	53894.000000	53894.000000	53894.000000	53894.000000	53894.000000	53894.000000	53894.000000	53894.000000	53894.000000	53894.000000	7073.000000	53894.000000	6964.000000	6167.000000	53894.000000	53894.000000	53894.000000	53894.000000	53894.000000	53894.000000	24073.000000	53894.000000	53894.000000	53894.000000	53894.000000	53894.000000	53894.000000	53894.000000	53894.000000	53894.000000	53894.000000	44900.000000	53894.000000	44645.000000	44900.000000	44900.000000	44900.000000	53894.000000	53894.000000	71934.000000	71934.000000	71934.000000	71934.000000
mean	8.737044	2.296130	3.642761	26.230940	0.164292	0.086967	0.243721	1.604400	0.036870	0.017893	30.239725	11.923265	1.263413	1.476127	0.006952	0.916621	0.101619	2.807974	12.122593	0.088401	0.042514	0.019005	0.654047	-0.321600	0.003197	2.308716	3.553782	26.410728	0.163935	0.024842	0.013303	29.696497	11.972486	1.276131	0.784288	0.006089	0.835079	0.128040	0.127585	0.281369	0.004446	9.167284	0.273487	0.726513	0.403133	0.250785
std	8.007133	6.787657	10.659336	78.084084	0.582966	0.371074	0.846369	5.800552	0.206199	0.133263	90.500488	36.458935	3.791816	8.808963	0.080247	0.613608	0.102616	5.295317	25.253063	0.313124	0.197482	0.126039	1.356862	2.442123	0.054019	2.282754	3.305519	31.217102	0.393968	0.146133	0.105516	41.098030	15.390478	1.536639	3.112586	0.071068	4.459200	0.088436	0.146611	0.220833	0.059074	7.706216	0.445752	0.445752	0.490531	0.433468
min	-6.960000	0.000000	0.000000	-6.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	-13.000000	-2.000000	0.000000	-37.137157	0.000000	-0.083333	-0.180825	0.000000	-26.000000	0.000000	0.000000	0.000000	0.000000	-17.034384	0.000000	0.000000	0.000000	-16.000000	0.000000	0.000000	0.000000	-51.000000	-8.000000	0.000000	-23.140028	0.000000	-95.000000	0.014706	-1.800000	-0.635000	0.000000	-3.760000	0.000000	0.000000	0.000000	0.000000
25%	2.300000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	-3.354257	0.000000	0.681740	0.023866	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	-1.380116	0.000000	0.500000	1.000000	1.625000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	-0.743872	0.000000	0.243243	0.055556	0.014125	0.104721	0.000000	3.000000	0.000000	0.000000	0.000000	0.000000
50%	6.600000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.761093	0.000000	0.874994	0.089666	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	-0.229281	0.000000	2.000000	3.000000	15.857142	0.000000	0.000000	0.000000	12.000000	7.000000	1.000000	0.421129	0.000000	0.785714	0.107143	0.080230	0.215183	0.000000	7.300000	0.000000	1.000000	0.000000	0.000000
75%	13.280000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	6.695111	0.000000	1.080228	0.167983	3.000000	11.000000	0.000000	0.000000	0.000000	1.000000	0.827914	0.000000	3.500000	5.461538	40.000000	0.000000	0.000000	0.000000	47.500000	17.000000	2.000000	2.110855	0.000000	1.333333	0.181818	0.210844	0.410248	0.000000	13.600000	1.000000	1.000000	1.000000	1.000000
max	59.500000	45.000000	68.000000	527.000000	7.000000	5.000000	11.000000	91.000000	4.000000	3.000000	662.000000	287.000000	29.000000	41.582783	2.000000	22.000000	0.582595	37.000000	238.000000	6.000000	4.000000	2.000000	14.000000	15.164796	2.000000	18.000000	23.000000	300.000000	4.000000	2.000000	2.000000	314.000000	156.000000	13.000000	25.490833	2.000000	150.000000	0.590909	1.842105	1.789474	1.000000	57.900002	1.000000	1.000000	1.000000	1.000000

	player_id	player_name	game_number	position	team	season	week	opponent_team	fantasy_points_ppr	day_slate	location	completions_lagged	attempts_lagged	passing_yards_lagged	passing_tds_lagged	interceptions_lagged	sacks_lagged	sack_yards_lagged	sack_fumbles_lagged	sack_fumbles_lost_lagged	passing_air_yards_lagged	passing_yards_after_catch_lagged	passing_first_downs_lagged	passing_epa_lagged	passing_2pt_conversions_lagged	pacr_lagged	dakota_lagged	carries_lagged	rushing_yards_lagged	rushing_tds_lagged	rushing_fumbles_lagged	rushing_fumbles_lost_lagged	rushing_first_downs_lagged	rushing_epa_lagged	rushing_2pt_conversions_lagged	fantasy_points_ppr_lagged	used_opponent_avg	used_game_number_avg	DNP_prior_season_game	never_seen
0	00-0004161	Jake Delhomme	1.0	QB	CLE	2010	1	TB	9.180000	Sunday_Morning	TB	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	0	1	1	1
3	00-0005106	Brett Favre	1.0	QB	MIN	2010	1	NO	8.840000	Thursday_Night	NO	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	0	1	1	1
7	00-0007091	Matt Hasselbeck	1.0	QB	SEA	2010	1	SF	19.200001	Sunday_Afternoon	SEA	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	0	1	1	1
9	00-0010346	Peyton Manning	1.0	QB	IND	2010	1	HOU	29.320000	Sunday_Morning	HOU	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	0	1	1	1
11	00-0011022	Donovan McNabb	1.0	QB	WAS	2010	1	DAL	8.540000	Sunday_Night	WAS	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	0	1	1	1
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
71954	00-0039732	Bo Nix	16.0	QB	DEN	2024	17	CIN	21.860001	Saturday_Afternoon	CIN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	0	1	1	1
71963	00-0039851	Drake Maye	16.0	QB	NE	2024	17	LAC	9.880000	Saturday_Morning	NE	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	0	1	1	1
71974	00-0039910	Jayden Daniels	16.0	QB	WAS	2024	17	ATL	31.780001	Sunday_Night	WAS	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	0	1	1	1
71977	00-0039917	Michael Penix	16.0	QB	ATL	2024	17	WAS	11.220000	Sunday_Night	WAS	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	0	1	1	1
71978	00-0039918	Caleb Williams	16.0	QB	CHI	2024	17	SEA	6.580000	Thursday_Night	CHI	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	0	1	1	1

	fantasy_points_ppr	completions_lagged	attempts_lagged	passing_yards_lagged	passing_tds_lagged	interceptions_lagged	sacks_lagged	sack_yards_lagged	sack_fumbles_lagged	sack_fumbles_lost_lagged	passing_air_yards_lagged	passing_yards_after_catch_lagged	passing_first_downs_lagged	passing_epa_lagged	passing_2pt_conversions_lagged	pacr_lagged	dakota_lagged	carries_lagged	rushing_yards_lagged	rushing_tds_lagged	rushing_fumbles_lagged	rushing_fumbles_lost_lagged	rushing_first_downs_lagged	rushing_epa_lagged	rushing_2pt_conversions_lagged	fantasy_points_ppr_lagged	used_opponent_avg	used_game_number_avg	DNP_prior_season_game	never_seen
count	8632.000000	6691.000000	6691.000000	6691.000000	6691.000000	6691.000000	6691.000000	6691.000000	6691.000000	6691.000000	6691.000000	6691.000000	6691.000000	6483.000000	6691.000000	6469.000000	6137.000000	6691.000000	6691.000000	6691.000000	6691.000000	6691.000000	6691.000000	5961.000000	6691.000000	6691.000000	8632.000000	8632.000000	8632.000000	8632.000000
mean	13.883455	18.447055	29.235905	210.449341	1.313496	0.694083	1.949106	12.840508	0.295789	0.143424	242.102234	95.711441	10.140048	1.615741	0.055498	0.939553	0.102088	2.981317	12.583021	0.132775	0.164132	0.047891	0.950307	0.085730	0.007497	14.121841	0.288230	0.711770	0.382646	0.224861
std	8.965117	8.544700	12.890738	101.453125	1.104650	0.823712	1.554646	11.186320	0.514429	0.352879	120.482422	51.751907	5.071029	9.144935	0.220754	0.591098	0.102339	2.580728	18.018902	0.350180	0.376116	0.197346	1.215730	2.500479	0.079886	8.454957	0.452965	0.452965	0.486061	0.417515
min	-6.960000	0.000000	0.000000	-2.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	-13.000000	-2.000000	0.000000	-37.137157	0.000000	-0.083333	-0.180825	0.000000	-15.000000	0.000000	0.000000	0.000000	0.000000	-16.318796	0.000000	-3.760000	0.000000	0.000000	0.000000	0.000000
25%	7.300000	14.000000	24.000000	158.000000	0.333333	0.000000	1.000000	4.000000	0.000000	0.000000	176.000000	64.000000	7.000000	-3.755118	0.000000	0.699764	0.024158	1.000000	0.000000	0.000000	0.000000	0.000000	0.000000	-1.119732	0.000000	8.340000	0.000000	0.000000	0.000000	0.000000
50%	13.950000	20.000000	31.000000	222.000000	1.000000	0.500000	2.000000	11.000000	0.000000	0.000000	251.000000	97.000000	10.833333	1.054443	0.000000	0.880682	0.089735	2.333333	6.000000	0.000000	0.000000	0.000000	0.750000	0.000000	0.000000	14.100000	0.000000	1.000000	0.000000	0.000000
75%	19.820000	24.000000	38.000000	278.000000	2.000000	1.000000	3.000000	19.000000	0.500000	0.000000	319.000000	129.000000	13.500000	7.439728	0.000000	1.081380	0.168449	4.000000	18.000000	0.000000	0.000000	0.000000	1.142857	1.276356	0.000000	19.600000	1.000000	1.000000	1.000000	0.000000
max	51.880001	45.000000	68.000000	527.000000	7.000000	5.000000	11.000000	91.000000	4.000000	3.000000	662.000000	287.000000	29.000000	41.582783	2.000000	22.000000	0.582595	26.000000	157.000000	3.000000	4.000000	2.000000	10.000000	15.164796	1.000000	49.320000	1.000000	1.000000	1.000000	1.000000

	player_id	player_name	player_display_name	position	position_group	headshot_url	team	season	week	season_type	opponent_team	completions	attempts	passing_yards	passing_tds	interceptions	sacks	sack_yards	sack_fumbles	sack_fumbles_lost	passing_air_yards	passing_yards_after_catch	passing_first_downs	passing_epa	passing_2pt_conversions	pacr	dakota	carries	rushing_yards	rushing_tds	rushing_fumbles	rushing_fumbles_lost	rushing_first_downs	rushing_epa	rushing_2pt_conversions	receptions	targets	receiving_yards	receiving_tds	receiving_fumbles	receiving_fumbles_lost	receiving_air_yards	receiving_yards_after_catch	receiving_first_downs	receiving_epa	receiving_2pt_conversions	racr	target_share	air_yards_share	wopr	special_teams_tds	fantasy_points	fantasy_points_ppr
0	00-0000865	None	Charlie Batch	QB	QB	None	PIT	2010	2	REG	TEN	5	11	25.0	0	0.0	2.0	6.0	0	0	107.0	24.0	0.0	-10.005162	0	0.233645	-0.004243	1	0.0	0	1.0	0.0	0.0	-0.982634	0	0	0	0.0	0	0.0	0.0	0.0	0.0	0.0	NaN	0	NaN	NaN	NaN	NaN	0.0	1.000000	1.000000
1	00-0000865	None	Charlie Batch	QB	QB	None	PIT	2010	3	REG	TB	12	17	186.0	3	2.0	0.0	-0.0	0	0	153.0	55.0	7.0	7.614506	0	1.215686	0.332648	5	26.0	0	0.0	0.0	2.0	3.388114	0	0	0	0.0	0	0.0	0.0	0.0	0.0	0.0	NaN	0	NaN	NaN	NaN	NaN	0.0	18.040001	18.040001
2	00-0000865	None	Charlie Batch	QB	QB	None	PIT	2010	4	REG	BAL	12	21	141.0	0	1.0	2.0	15.0	0	0	278.0	46.0	6.0	-0.152696	0	0.507194	0.053022	1	4.0	0	0.0	0.0	0.0	-0.714745	0	0	0	0.0	0	0.0	0.0	0.0	0.0	0.0	NaN	0	NaN	NaN	NaN	NaN	0.0	4.040000	4.040000
3	00-0001231	None	Josh Bidwell	P	SPEC	None	WAS	2010	1	REG	DAL	0	0	0.0	0	0.0	0.0	0.0	0	0	0.0	0.0	0.0	NaN	0	NaN	NaN	1	0.0	0	1.0	0.0	0.0	-3.118365	0	0	0	0.0	0	0.0	0.0	0.0	0.0	0.0	NaN	0	NaN	NaN	NaN	NaN	0.0	0.000000	0.000000
4	00-0001545	None	Todd Bouman	QB	QB	None	JAX	2010	7	REG	KC	18	34	222.0	2	2.0	1.0	6.0	0	0	250.0	121.0	12.0	-5.145036	0	0.888000	0.079166	2	8.0	0	0.0	0.0	1.0	2.377490	0	0	0	0.0	0	0.0	0.0	0.0	0.0	0.0	NaN	0	NaN	NaN	NaN	NaN	0.0	13.680000	13.680000
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
5592	00-0039921	T.Benson	Trey Benson	RB	RB	https://static.www.nfl.com/image/upload/f_auto...	ARI	2024	10	REG	NYJ	0	0	0.0	0	0.0	0.0	0.0	0	0	0.0	0.0	0.0	NaN	0	NaN	NaN	10	62.0	0	0.0	0.0	3.0	1.512646	0	2	2	25.0	0	0.0	0.0	-8.0	33.0	2.0	1.987017	0	-3.125	0.083333	-0.072072	0.074550	0.0	8.700000	10.700000
5593	00-0039921	T.Benson	Trey Benson	RB	RB	https://static.www.nfl.com/image/upload/f_auto...	ARI	2024	12	REG	SEA	0	0	0.0	0	0.0	0.0	0.0	0	0	0.0	0.0	0.0	NaN	0	NaN	NaN	4	18.0	0	0.0	0.0	0.0	-0.326119	0	0	0	0.0	0	0.0	0.0	0.0	0.0	0.0	NaN	0	NaN	NaN	NaN	NaN	0.0	1.800000	1.800000
5594	00-0039921	T.Benson	Trey Benson	RB	RB	https://static.www.nfl.com/image/upload/f_auto...	ARI	2024	13	REG	MIN	0	0	0.0	0	0.0	0.0	0.0	0	0	0.0	0.0	0.0	NaN	0	NaN	NaN	3	20.0	0	0.0	0.0	1.0	0.515926	0	0	0	0.0	0	0.0	0.0	0.0	0.0	0.0	NaN	0	NaN	NaN	NaN	NaN	0.0	2.000000	2.000000
5595	00-0039921	T.Benson	Trey Benson	RB	RB	https://static.www.nfl.com/image/upload/f_auto...	ARI	2024	14	REG	SEA	0	0	0.0	0	0.0	0.0	0.0	0	0	0.0	0.0	0.0	NaN	0	NaN	NaN	2	15.0	0	0.0	0.0	1.0	0.053110	0	1	1	4.0	0	0.0	0.0	-5.0	9.0	0.0	-0.144496	0	-0.800	0.026316	-0.031847	0.017181	0.0	1.900000	2.900000
5596	00-0039921	T.Benson	Trey Benson	RB	RB	https://static.www.nfl.com/image/upload/f_auto...	ARI	2024	15	REG	NE	0	0	0.0	0	0.0	0.0	0.0	0	0	0.0	0.0	0.0	NaN	0	NaN	NaN	5	22.0	0	0.0	0.0	2.0	-0.384363	0	0	0	0.0	0	0.0	0.0	0.0	0.0	0.0	NaN	0	NaN	NaN	NaN	NaN	0.0	2.200000	2.200000

	fantasy_points_ppr	receptions_lagged	targets_lagged	receiving_yards_lagged	receiving_tds_lagged	receiving_fumbles_lagged	receiving_fumbles_lost_lagged	receiving_air_yards_lagged	receiving_yards_after_catch_lagged	receiving_first_downs_lagged	receiving_epa_lagged	receiving_2pt_conversions_lagged	racr_lagged	target_share_lagged	air_yards_share_lagged	wopr_lagged	special_teams_tds_lagged	fantasy_points_ppr_lagged	used_opponent_avg	used_game_number_avg	DNP_prior_season_game	never_seen
count	28345.000000	21130.000000	21130.000000	21130.000000	21130.000000	21130.000000	21130.00000	21130.000000	21130.000000	21130.000000	20904.000000	21130.000000	20873.000000	20904.000000	20904.000000	20904.000000	21130.000000	21130.000000	28345.000000	28345.000000	28345.000000	28345.000000
mean	8.762276	3.273661	5.363094	42.733273	0.257692	0.033653	0.01840	59.632870	14.324631	2.050847	1.230115	0.009542	0.936981	0.160992	0.215151	0.392093	0.009029	9.290085	0.279097	0.720903	0.392556	0.254542
std	7.890453	2.445860	3.486932	36.228554	0.479321	0.168519	0.12339	46.184597	15.892762	1.739775	3.707499	0.089583	1.739763	0.096199	0.151077	0.237627	0.084114	7.597829	0.448563	0.448563	0.488328	0.435611
min	-3.000000	0.000000	0.000000	-7.000000	0.000000	0.000000	0.00000	-32.000000	-8.000000	0.000000	-23.140028	0.000000	-2.000000	0.015873	-0.220690	0.009724	0.000000	-2.780000	0.000000	0.000000	0.000000	0.000000
25%	2.500000	1.000000	2.500000	15.000000	0.000000	0.000000	0.00000	24.000000	3.000000	1.000000	-0.785849	0.000000	0.433962	0.081081	0.096059	0.195288	0.000000	3.400000	0.000000	0.000000	0.000000	0.000000
50%	6.700000	3.000000	5.000000	35.000000	0.000000	0.000000	0.00000	51.000000	9.500000	2.000000	0.865888	0.000000	0.743902	0.150485	0.195275	0.363758	0.000000	7.600000	0.000000	1.000000	0.000000	0.000000
75%	13.000000	5.000000	7.631119	62.000000	0.333333	0.000000	0.00000	86.000000	20.000000	3.000000	3.118169	0.000000	1.100000	0.225186	0.310475	0.549950	0.000000	13.400000	1.000000	1.000000	1.000000	1.000000
max	57.900002	18.000000	23.000000	300.000000	4.000000	2.000000	2.00000	314.000000	153.000000	13.000000	25.490833	2.000000	116.000000	0.590909	1.842105	1.789474	1.000000	57.900002	1.000000	1.000000	1.000000	1.000000

	fantasy_points_ppr	carries_lagged	rushing_yards_lagged	rushing_tds_lagged	rushing_fumbles_lagged	rushing_fumbles_lost_lagged	rushing_first_downs_lagged	rushing_epa_lagged	rushing_2pt_conversions_lagged	receptions_lagged	targets_lagged	receiving_yards_lagged	receiving_tds_lagged	receiving_fumbles_lagged	receiving_fumbles_lost_lagged	receiving_air_yards_lagged	receiving_yards_after_catch_lagged	receiving_first_downs_lagged	receiving_epa_lagged	receiving_2pt_conversions_lagged	racr_lagged	target_share_lagged	air_yards_share_lagged	wopr_lagged	special_teams_tds_lagged	fantasy_points_ppr_lagged	used_opponent_avg	used_game_number_avg	DNP_prior_season_game	never_seen
count	20533.000000	15016.000000	15016.000000	15016.000000	15016.000000	15016.000000	15016.000000	13710.000000	15016.000000	15016.000000	15016.000000	15016.000000	15016.000000	15016.000000	15016.000000	15016.000000	15016.000000	15016.000000	12782.000000	15016.000000	12647.000000	12782.000000	12782.000000	12782.000000	15016.000000	15016.000000	20533.000000	20533.000000	20533.000000	20533.000000
mean	8.228588	8.441637	36.051937	0.245504	0.074314	0.044096	1.824823	-0.639104	0.007589	1.841406	2.450811	14.432546	0.067671	0.021699	0.010202	1.961093	14.140253	0.652019	0.084189	0.001610	0.240903	0.089574	0.006832	0.139143	0.003036	8.702289	0.263722	0.736278	0.423659	0.268689
std	7.904483	7.071124	35.454884	0.495610	0.248602	0.189242	1.904732	2.641750	0.083962	1.846477	2.266974	17.769648	0.243754	0.137264	0.092378	9.221778	17.246443	0.895203	2.020171	0.034455	7.884886	0.059868	0.045382	0.098151	0.048632	7.556914	0.440661	0.440661	0.494150	0.443289
min	-2.800000	0.000000	-11.000000	0.000000	0.000000	0.000000	0.000000	-17.034384	0.000000	0.000000	0.000000	-13.000000	0.000000	0.000000	0.000000	-51.000000	-5.000000	0.000000	-15.334975	0.000000	-95.000000	0.014706	-1.800000	-0.635000	0.000000	-2.600000	0.000000	0.000000	0.000000	0.000000
25%	2.000000	2.000000	7.000000	0.000000	0.000000	0.000000	0.000000	-1.846540	0.000000	0.500000	1.000000	1.000000	0.000000	0.000000	0.000000	-2.000000	0.500000	0.000000	-0.816950	0.000000	-1.538822	0.041667	-0.011959	0.063677	0.000000	2.700000	0.000000	0.000000	0.000000	0.000000
50%	5.900000	7.000000	26.000000	0.000000	0.000000	0.000000	1.000000	-0.533364	0.000000	1.000000	2.000000	9.000000	0.000000	0.000000	0.000000	0.000000	8.500000	0.200000	-0.041459	0.000000	0.142857	0.074074	0.003685	0.113982	0.000000	6.650000	0.000000	1.000000	0.000000	0.000000
75%	12.400000	13.000000	56.000000	0.200000	0.000000	0.000000	3.000000	0.596231	0.000000	3.000000	3.600000	21.000000	0.000000	0.000000	0.000000	4.000000	21.000000	1.000000	0.986842	0.000000	2.200000	0.121951	0.020198	0.187557	0.000000	12.950000	1.000000	1.000000	1.000000	1.000000
max	59.500000	37.000000	238.000000	6.000000	2.000000	2.000000	14.000000	14.181790	2.000000	15.000000	20.000000	156.000000	3.000000	2.000000	1.000000	149.000000	156.000000	7.000000	13.652180	1.000000	150.000000	0.428571	0.518182	1.005584	1.000000	56.200001	1.000000	1.000000	1.000000	1.000000

	player_id	player_name	game_number	position	team	season	week	opponent_team	fantasy_points_ppr	day_slate	location	receptions_lagged	targets_lagged	receiving_yards_lagged	receiving_tds_lagged	receiving_fumbles_lagged	receiving_fumbles_lost_lagged	receiving_air_yards_lagged	receiving_yards_after_catch_lagged	receiving_first_downs_lagged	receiving_epa_lagged	receiving_2pt_conversions_lagged	racr_lagged	target_share_lagged	air_yards_share_lagged	wopr_lagged	special_teams_tds_lagged	fantasy_points_ppr_lagged	used_opponent_avg	used_game_number_avg	DNP_prior_season_game	never_seen
6	00-0006101	Tony Gonzalez	1.0	TE	ATL	2010	1	PIT	5.5	Sunday_Morning	PIT	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	0	1	1	1
8	00-0009323	Jimmy Kleinsasser	1.0	TE	MIN	2010	1	NO	1.6	Thursday_Night	NO	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	0	1	1	1
15	00-0014655	Mike Sellers	1.0	TE	WAS	2010	1	DAL	1.2	Sunday_Night	WAS	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	0	1	1	1
33	00-0020516	Todd Heap	1.0	TE	BAL	2010	1	NYJ	13.2	Monday_Afternoon	NYJ	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	0	1	1	1
37	00-0021148	Jeremy Shockey	1.0	TE	NO	2010	1	MIN	5.3	Thursday_Night	NO	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	0	1	1	1
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
71938	00-0039338	Brock Bowers	16.0	TE	LV	2024	17	NO	14.6	Sunday_Morning	NO	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	0	1	1	1
71940	00-0039356	Ja'Tavion Sanders	16.0	TE	CAR	2024	17	TB	1.5	Sunday_Morning	TB	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	0	1	1	1
71941	00-0039359	Cade Stover	16.0	TE	HOU	2024	17	BAL	1.1	Wednesday_Afternoon	HOU	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	0	1	1	1
71953	00-0039631	Cam Grandy	16.0	TE	CIN	2024	17	DEN	1.7	Saturday_Afternoon	CIN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	0	1	1	1
71957	00-0039754	Devin Culp	16.0	TE	TB	2024	17	CAR	8.2	Sunday_Morning	TB	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	0	1	1	1

	fantasy_points_ppr	receptions_lagged	targets_lagged	receiving_yards_lagged	receiving_tds_lagged	receiving_fumbles_lagged	receiving_fumbles_lost_lagged	receiving_air_yards_lagged	receiving_yards_after_catch_lagged	receiving_first_downs_lagged	receiving_epa_lagged	receiving_2pt_conversions_lagged	racr_lagged	target_share_lagged	air_yards_share_lagged	wopr_lagged	special_teams_tds_lagged	fantasy_points_ppr_lagged	used_opponent_avg	used_game_number_avg	DNP_prior_season_game	never_seen
count	14424.000000	11057.000000	11057.000000	11057.000000	11057.000000	11057.000000	11057.000000	11057.000000	11057.000000	11057.000000	11001.000000	11057.000000	10917.000000	11001.000000	11001.000000	11001.000000	11057.000000	11057.000000	14424.000000	14424.000000	14424.000000	14424.000000
mean	6.331412	2.486529	3.729680	27.406677	0.213662	0.027169	0.015826	28.060663	11.730103	1.411277	0.770623	0.008907	1.335781	0.111888	0.103755	0.240462	0.000098	6.565917	0.267540	0.732460	0.406961	0.233430
std	6.193554	2.008066	2.743930	26.212807	0.433783	0.153584	0.115800	27.342024	12.830843	1.401027	2.768183	0.085473	1.676981	0.077169	0.097868	0.176271	0.004820	6.033113	0.442692	0.442692	0.491284	0.423028
min	-2.000000	0.000000	0.000000	-6.000000	0.000000	0.000000	0.000000	-13.000000	-7.000000	0.000000	-15.294791	0.000000	-2.500000	0.015873	-0.121495	0.007336	0.000000	-2.000000	0.000000	0.000000	0.000000	0.000000
25%	1.800000	1.000000	1.571429	8.875000	0.000000	0.000000	0.000000	8.000000	3.000000	0.384615	-0.551518	0.000000	0.583333	0.048780	0.029777	0.098117	0.000000	2.150000	0.000000	0.000000	0.000000	0.000000
50%	4.300000	2.000000	3.000000	20.000000	0.000000	0.000000	0.000000	19.571428	8.000000	1.000000	0.540752	0.000000	1.014925	0.090909	0.075000	0.191595	0.000000	4.700000	0.000000	1.000000	0.000000	0.000000
75%	9.200000	3.500000	5.000000	39.000000	0.166667	0.000000	0.000000	41.000000	16.500000	2.000000	1.966459	0.000000	1.514490	0.159716	0.150704	0.344308	0.000000	9.200000	1.000000	1.000000	1.000000	0.000000
max	45.000000	15.000000	17.000000	210.000000	4.000000	2.000000	2.000000	208.000000	138.000000	11.000000	17.672009	1.000000	31.500000	0.481481	0.825397	1.129885	0.250000	41.900002	1.000000	1.000000	1.000000	1.000000

	player_id	player_name	game_number	position	team	season	week	opponent_team	day_slate	location	completions_lagged	attempts_lagged	passing_yards_lagged	passing_tds_lagged	interceptions_lagged	sacks_lagged	sack_yards_lagged	sack_fumbles_lagged	sack_fumbles_lost_lagged	passing_air_yards_lagged	passing_yards_after_catch_lagged	passing_first_downs_lagged	passing_epa_lagged	passing_2pt_conversions_lagged	pacr_lagged	dakota_lagged	carries_lagged	rushing_yards_lagged	rushing_tds_lagged	rushing_fumbles_lagged	rushing_fumbles_lost_lagged	rushing_first_downs_lagged	rushing_epa_lagged	rushing_2pt_conversions_lagged	fantasy_points_ppr_lagged	used_opponent_avg	used_game_number_avg	DNP_prior_season_game	never_seen	prediction	fantasy_points_ppr
60886.0	00-0034869	Sam Darnold	13.0	QB	CAR	2022	14	SEA	Sunday_Afternoon	SEA	19.454545	33.090909	209.818176	0.636364	1.090909	3.000000	17.454546	0.636364	0.181818	242.000000	112.181816	10.272727	-6.573395	0.090909	0.932098	0.040188	3.636364	18.454546	0.454545	0.090909	0.090909	1.909091	1.354638	0.0	12.965454	0	1	1	0	13.161040	11.80
27261.0	00-0024296	Charlie Whitehurst	13.0	QB	IND	2015	14	JAX	Sunday_Morning	JAX	20.500000	31.500000	260.000000	0.500000	0.000000	3.500000	19.000000	0.500000	0.000000	246.500000	97.000000	11.000000	2.992638	0.000000	1.081960	0.096804	2.500000	6.500000	0.000000	0.000000	0.000000	0.500000	-0.094627	0.0	13.050000	1	0	0	0	3.699607	-1.48
20845.0	00-0024226	Jay Cutler	8.0	QB	CHI	2014	8	NE	Sunday_Morning	NE	20.900000	33.100000	239.500000	1.700000	1.100000	1.800000	13.000000	0.400000	0.300000	296.799988	82.599998	11.700000	0.062747	0.300000	0.809315	0.112973	2.300000	11.800000	0.000000	0.100000	0.000000	0.700000	0.476893	0.0	15.360001	0	1	1	0	15.020247	21.68
45987.0	00-0034869	Sam Darnold	12.0	QB	NYJ	2019	13	CIN	Sunday_Morning	CIN	18.583333	32.166667	224.833328	1.416667	1.250000	2.166667	14.333333	0.000000	0.000000	293.833344	103.083336	10.166667	-0.256866	0.000000	0.829197	0.075765	3.500000	8.916667	0.083333	0.333333	0.083333	1.250000	-0.831336	0.0	13.384999	0	1	1	0	13.287729	10.46
40342.0	00-0030533	Matt Barkley	10.0	QB	BUF	2018	10	NYJ	Sunday_Morning	NYJ	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	0	1	1	1	2.860645	21.08
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
63301.0	00-0034857	Josh Allen	5.0	QB	BUF	2023	5	JAX	Sunday_global	global	20.000000	31.000000	424.000000	4.000000	1.000000	0.000000	0.000000	0.000000	0.000000	420.000000	202.000000	15.000000	24.854641	0.000000	1.009524	0.330383	5.000000	42.000000	0.000000	0.000000	0.000000	2.000000	2.863234	0.0	35.160000	0	1	0	0	18.023715	27.76
3928.0	00-0026993	Josh Freeman	13.0	QB	TB	2010	14	WAS	Sunday_Morning	WAS	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	0	1	1	1	15.452269	15.64
57741.0	00-0035710	Daniel Jones	3.0	QB	NYG	2022	3	DAL	Monday_Night	NYG	5.000000	13.000000	98.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	192.000000	36.000000	4.000000	0.887441	0.000000	0.510417	-0.051411	3.000000	9.000000	0.000000	0.000000	0.000000	1.000000	-0.780850	0.0	4.820000	1	0	0	0	15.084281	13.74
63952.0	00-0037077	Sam Howell	7.0	QB	WAS	2023	7	NYG	Sunday_Morning	NYG	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	0	1	1	1	13.215383	9.46
59319.0	00-0019596	Tom Brady	9.0	QB	TB	2022	9	LA	Sunday_Afternoon	TB	41.000000	55.000000	432.000000	1.000000	0.000000	3.000000	21.000000	1.000000	0.000000	382.000000	238.000000	24.000000	13.163302	0.000000	1.130890	0.137871	3.000000	14.000000	1.000000	0.000000	0.000000	1.000000	0.562927	0.0	28.680000	1	0	0	0	18.279850	15.20

	fantasy_points_ppr	never_seen	DNP_prior_season_game
count	1091.000000	1091.000000	1091.000000
mean	109.846001	0.292392	0.525917
std	111.257225	0.455070	0.436633
min	-7.280000	0.000000	0.000000
25%	8.280000	0.000000	0.000000
50%	68.160004	0.000000	0.500000
75%	206.639999	1.000000	1.000000
max	415.679993	1.000000	1.000000

	player_id	player_name	game_number	position	team	season	week	opponent_team	fantasy_points_ppr	day_slate	location	completions_lagged	attempts_lagged	passing_yards_lagged	passing_tds_lagged	interceptions_lagged	sacks_lagged	sack_yards_lagged	sack_fumbles_lagged	sack_fumbles_lost_lagged	passing_air_yards_lagged	passing_yards_after_catch_lagged	passing_first_downs_lagged	passing_epa_lagged	passing_2pt_conversions_lagged	pacr_lagged	dakota_lagged	carries_lagged	rushing_yards_lagged	rushing_tds_lagged	rushing_fumbles_lagged	rushing_fumbles_lost_lagged	rushing_first_downs_lagged	rushing_epa_lagged	rushing_2pt_conversions_lagged	fantasy_points_ppr_lagged	used_opponent_avg	used_game_number_avg	DNP_prior_season_game	never_seen
0	00-0007091	Matt Hasselbeck	1.0	QB	TEN	2011	1	JAX	17.420000	Sunday_Morning	JAX	18.0	23.0	170.00	2.00	1.00	1.0	5.0	0.0	0.0	132.00	80.00	9.0	6.027418	0.0	1.287879	0.232092	2.00	4.0	1.0	0.0	0.0	1.00	0.632133	0.0	19.200001	0	1	0	0
1	00-0007091	Matt Hasselbeck	2.0	QB	TEN	2011	2	BAL	16.320000	Sunday_Morning	TEN	20.0	35.0	233.00	1.00	3.00	1.0	3.0	0.0	0.0	335.00	112.00	11.0	1.475178	0.0	0.695522	0.090619	3.00	21.0	1.0	0.0	0.0	2.00	5.171046	0.0	15.420000	0	1	0	0
2	00-0007091	Matt Hasselbeck	3.0	QB	TEN	2011	3	DEN	18.240000	Sunday_Morning	TEN	20.0	35.0	233.00	1.00	3.00	1.0	3.0	0.0	0.0	335.00	112.00	11.0	1.475178	0.0	0.695522	0.090619	3.00	21.0	1.0	0.0	0.0	2.00	5.171046	0.0	15.420000	1	0	0	0
3	00-0007091	Matt Hasselbeck	4.0	QB	TEN	2011	4	CLE	19.299999	Sunday_Morning	CLE	20.0	36.0	191.00	0.00	1.00	4.0	26.0	1.0	1.0	295.00	51.00	9.0	-18.075758	0.0	0.647458	-0.008021	1.00	3.0	0.0	0.0	0.0	0.00	-1.319311	0.0	3.940000	0	1	0	0
4	00-0007091	Matt Hasselbeck	5.0	QB	TEN	2011	5	PIT	12.780000	Sunday_Morning	PIT	25.0	40.0	242.00	1.00	0.00	0.0	0.0	0.0	0.0	270.00	69.00	14.0	2.947018	0.0	0.896296	0.070844	3.00	-3.0	0.0	0.0	0.0	0.00	-1.259380	0.0	13.380000	0	1	0	0
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
5553	00-0035704	Drew Lock	11.0	QB	NYG	2024	12	TB	0.120000	Sunday_Morning	NYG	12.0	19.0	135.75	0.75	0.75	1.5	6.5	0.0	0.0	155.25	76.25	5.0	-0.163431	0.0	0.820123	0.025158	1.25	3.5	0.0	0.0	0.0	0.25	-0.070308	0.0	7.280000	0	1	1	0
5554	00-0035704	Drew Lock	12.0	QB	NYG	2024	13	DAL	14.820000	Thursday_Afternoon	DAL	12.0	19.0	135.75	0.75	0.75	1.5	6.5	0.0	0.0	155.25	76.25	5.0	-0.163431	0.0	0.820123	0.025158	1.25	3.5	0.0	0.0	0.0	0.25	-0.070308	0.0	7.280000	0	1	1	0
5555	00-0035704	Drew Lock	13.0	QB	NYG	2024	14	NO	14.980000	Sunday_Morning	NYG	22.0	31.0	269.00	2.00	2.00	4.0	15.0	0.0	0.0	235.00	167.00	11.0	-0.236313	0.0	1.144681	0.078055	3.00	5.0	0.0	0.0	0.0	0.00	-1.308873	0.0	15.260000	0	1	0	0
5556	00-0035704	Drew Lock	15.0	QB	NYG	2024	16	ATL	6.500000	Sunday_Morning	ATL	12.0	19.0	135.75	0.75	0.75	1.5	6.5	0.0	0.0	155.25	76.25	5.0	-0.163431	0.0	0.820123	0.025158	1.25	3.5	0.0	0.0	0.0	0.25	-0.070308	0.0	7.280000	0	1	1	0
5557	00-0035704	Drew Lock	16.0	QB	NYG	2024	17	IND	34.360001	Sunday_Morning	NYG	12.0	19.0	135.75	0.75	0.75	1.5	6.5	0.0	0.0	155.25	76.25	5.0	-0.163431	0.0	0.820123	0.025158	1.25	3.5	0.0	0.0	0.0	0.25	-0.070308	0.0	7.280000	0	1	1	0

	player_id	player_name	game_number	position	team	season	week	opponent_team	day_slate	location	completions_lagged	attempts_lagged	passing_yards_lagged	passing_tds_lagged	interceptions_lagged	sacks_lagged	sack_yards_lagged	sack_fumbles_lagged	sack_fumbles_lost_lagged	passing_air_yards_lagged	passing_yards_after_catch_lagged	passing_first_downs_lagged	passing_epa_lagged	passing_2pt_conversions_lagged	pacr_lagged	dakota_lagged	carries_lagged	rushing_yards_lagged	rushing_tds_lagged	rushing_fumbles_lagged	rushing_fumbles_lost_lagged	rushing_first_downs_lagged	rushing_epa_lagged	rushing_2pt_conversions_lagged	fantasy_points_ppr_lagged	used_opponent_avg	used_game_number_avg	DNP_prior_season_game	never_seen	prediction	fantasy_points_ppr
4669.0	00-0036971	Trevor Lawrence	7.0	QB	JAX	2022	7	NYG	Sunday_Morning	JAX	32.000000	54.000000	238.000000	1.000000	1.000000	1.0	11.000000	0.000000	0.000000	279.0	154.000000	17.000000	-7.699882	0.000000	0.853047	-0.010704	3.000000	11.000000	0.000000	0.000000	0.000000	0.000000	-1.292308	0.0	12.620000	0	1	0	0	14.209453	18.700001
1490.0	00-0030565	Geno Smith	1.0	QB	NYJ	2014	1	LV	Sunday_Morning	NYJ	16.000000	25.000000	219.000000	1.000000	1.000000	1.0	10.000000	0.000000	0.000000	131.0	123.000000	9.000000	4.915254	0.000000	1.671756	0.198605	5.000000	50.000000	1.000000	0.000000	0.000000	3.000000	6.536024	0.0	21.760000	1	0	0	0	15.429075	12.640000
412.0	00-0021429	Carson Palmer	2.0	QB	LV	2012	2	MIA	Sunday_Morning	MIA	20.000000	41.000000	273.000000	2.000000	1.000000	2.0	15.000000	0.000000	0.000000	437.0	88.000000	12.000000	-1.226834	0.000000	0.624714	0.002722	1.000000	2.000000	0.000000	0.000000	0.000000	1.000000	0.360314	0.0	17.120001	1	0	0	0	15.253438	17.620001
1721.0	00-0026158	Joe Flacco	1.0	QB	BAL	2015	1	DEN	Sunday_Afternoon	DEN	35.000000	62.000000	345.000000	1.000000	1.000000	3.0	16.000000	0.000000	0.000000	489.0	124.000000	16.000000	-4.871387	0.000000	0.705521	-0.023444	3.000000	7.000000	0.000000	0.000000	0.000000	2.000000	1.793058	0.0	16.500000	0	1	0	0	15.281090	0.680000
5349.0	00-0036212	Tua Tagovailoa	11.0	QB	MIA	2024	12	NE	Sunday_Morning	MIA	25.500000	37.500000	286.500000	2.000000	1.000000	1.5	8.500000	0.000000	0.000000	287.0	137.000000	16.000000	11.924225	0.000000	0.993973	0.159436	5.000000	5.500000	0.000000	1.500000	0.500000	0.500000	-5.464259	0.0	17.010000	1	0	0	0	15.815332	28.480000
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
5071.0	00-0037834	Brock Purdy	14.0	QB	SF	2023	15	ARI	Sunday_Afternoon	ARI	0.000000	0.000000	0.000000	0.000000	0.000000	0.0	0.000000	0.000000	0.000000	0.0	0.000000	0.000000	NaN	0.000000	NaN	NaN	3.000000	3.000000	0.000000	0.000000	0.000000	0.000000	0.314846	0.0	0.300000	1	0	0	0	17.403790	25.680000
44.0	00-0020531	Drew Brees	3.0	QB	NO	2011	3	HOU	Sunday_Morning	NO	30.000000	38.000000	365.000000	3.000000	2.000000	2.0	10.000000	0.000000	0.000000	281.0	206.000000	17.000000	8.612495	0.000000	1.298932	0.206769	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	NaN	0.0	22.600000	0	1	0	0	19.360854	26.700001
985.0	00-0027939	Cam Newton	10.0	QB	CAR	2013	11	NE	Monday_Night	CAR	16.000000	29.000000	252.000000	1.000000	0.000000	2.0	18.000000	0.000000	0.000000	291.0	115.000000	10.000000	6.545709	0.000000	0.865979	0.096505	11.000000	40.000000	0.000000	0.000000	0.000000	3.000000	0.216940	0.0	18.080000	0	1	0	0	17.970347	26.559999
674.0	00-0027946	Jake Locker	10.0	QB	TEN	2012	10	MIA	Sunday_Morning	MIA	9.000000	19.000000	140.000000	2.000000	0.000000	1.0	7.000000	0.000000	0.000000	191.0	42.000000	6.000000	9.031020	0.000000	0.732984	0.161103	1.000000	11.000000	0.000000	0.000000	0.000000	1.000000	1.875331	0.0	14.700000	0	1	0	0	12.935845	16.480000
4735.0	00-0034869	Sam Darnold	13.0	QB	CAR	2022	14	SEA	Sunday_Afternoon	SEA	19.454545	33.090909	209.818176	0.636364	1.090909	3.0	17.454546	0.636364	0.181818	242.0	112.181816	10.272727	-6.573395	0.090909	0.932098	0.040188	3.636364	18.454546	0.454545	0.090909	0.090909	1.909091	1.354638	0.0	12.965454	0	1	1	0	15.162555	11.800000

	player_id	player_name	game_number	position	team	season	week	opponent_team	day_slate	location	receptions_lagged	targets_lagged	receiving_yards_lagged	receiving_tds_lagged	receiving_fumbles_lagged	receiving_fumbles_lost_lagged	receiving_air_yards_lagged	receiving_yards_after_catch_lagged	receiving_first_downs_lagged	receiving_epa_lagged	receiving_2pt_conversions_lagged	racr_lagged	target_share_lagged	air_yards_share_lagged	wopr_lagged	special_teams_tds_lagged	fantasy_points_ppr_lagged	used_opponent_avg	used_game_number_avg	DNP_prior_season_game	never_seen	prediction	fantasy_points_ppr
37272.0	00-0033307	Kendrick Bourne	14.0	WR	SF	2017	15	TEN	Sunday_Afternoon	SF	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	0	1	1	1	4.368976	12.5
10162.0	00-0027681	Brandon LaFell	3.0	WR	CAR	2012	3	NYG	Thursday_Night	CAR	2.428571	3.857143	41.857143	0.214286	0.000000	0.000000	50.714287	16.142857	1.500000	1.954026	0.0	0.984467	0.125931	0.193018	0.324008	0.0	7.900000	0	1	1	0	9.580862	3.7
51852.0	00-0029000	Cole Beasley	1.0	WR	BUF	2021	1	PIT	Sunday_Morning	BUF	5.000000	10.000000	41.000000	0.000000	0.000000	0.000000	83.000000	26.000000	2.000000	-2.304453	0.0	0.493976	0.238095	0.182418	0.484835	0.0	9.100000	1	0	0	0	8.759766	14.0
23962.0	00-0030460	Markus Wheaton	2.0	WR	PIT	2015	2	SF	Sunday_Morning	PIT	5.000000	8.000000	38.000000	0.000000	0.000000	0.000000	45.000000	28.000000	2.000000	0.804461	0.0	0.844444	0.216216	0.202703	0.466216	0.0	11.000000	0	1	0	0	6.133824	8.7
10577.0	00-0029707	Justin Blackmon	4.0	WR	JAX	2012	4	CIN	Sunday_Afternoon	JAX	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	0	1	1	1	9.331889	10.8
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
66278.0	00-0037487	Deven Thompkins	14.0	WR	TB	2023	15	GB	Sunday_Morning	GB	1.000000	1.000000	7.000000	0.000000	0.000000	0.000000	4.000000	3.000000	0.000000	-0.060864	0.0	1.750000	0.023810	0.012903	0.044747	0.0	2.600000	0	1	0	0	2.446278	-0.1
20049.0	00-0027918	Andrew Hawkins	4.0	WR	CLE	2014	5	TEN	Sunday_Morning	TEN	1.285714	1.857143	17.857143	0.000000	0.000000	0.000000	8.714286	12.714286	0.571429	0.524039	0.0	2.672961	0.068853	0.032394	0.125955	0.0	3.114286	0	1	1	0	5.927177	5.6
63491.0	00-0030279	Keenan Allen	5.0	WR	LAC	2023	6	DAL	Monday_Night	LAC	6.444444	8.666667	72.222221	0.222222	0.111111	0.111111	75.444443	23.444445	3.444444	2.028409	0.0	1.198994	0.231277	0.293491	0.552360	0.0	14.866667	0	1	1	0	15.698590	21.5
60940.0	00-0036358	CeeDee Lamb	13.0	WR	DAL	2022	14	HOU	Sunday_Morning	DAL	7.000000	10.000000	61.000000	0.000000	0.000000	0.000000	60.000000	39.000000	4.000000	-0.434593	0.0	1.016667	0.256410	0.246914	0.557455	0.0	14.600000	0	1	0	0	15.806026	8.3
7375.0	00-0026937	Brandon Gibson	9.0	WR	LA	2011	10	CLE	Sunday_Morning	CLE	8.000000	9.000000	72.000000	0.000000	0.000000	0.000000	66.000000	16.000000	3.000000	4.935188	0.0	1.090909	0.214286	0.257812	0.501897	0.0	15.200000	0	1	0	0	8.234083	0.0

	player_id	player_name	game_number	position	team	season	week	opponent_team	fantasy_points_ppr	day_slate	location	carries_lagged	rushing_yards_lagged	rushing_tds_lagged	rushing_fumbles_lagged	rushing_fumbles_lost_lagged	rushing_first_downs_lagged	rushing_epa_lagged	rushing_2pt_conversions_lagged	receptions_lagged	targets_lagged	receiving_yards_lagged	receiving_tds_lagged	receiving_fumbles_lagged	receiving_fumbles_lost_lagged	receiving_air_yards_lagged	receiving_yards_after_catch_lagged	receiving_first_downs_lagged	receiving_epa_lagged	receiving_2pt_conversions_lagged	racr_lagged	target_share_lagged	air_yards_share_lagged	wopr_lagged	special_teams_tds_lagged	fantasy_points_ppr_lagged	used_opponent_avg	used_game_number_avg	DNP_prior_season_game	never_seen
35	00-0020536	LaDainian Tomlinson	1.0	RB	NYJ	2010	1	BAL	9.8	Monday_Afternoon	NYJ	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	0	1	1	1
56	00-0021755	Earnest Graham	1.0	RB	TB	2010	1	CLE	0.8	Sunday_Morning	TB	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	0	1	1	1
72	00-0022405	Lousaka Polite	1.0	FB	MIA	2010	1	BUF	3.8	Sunday_Morning	BUF	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	0	1	1	1
83	00-0022821	Michael Turner	1.0	RB	ATL	2010	1	PIT	5.9	Sunday_Morning	PIT	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	0	1	1	1
85	00-0022897	Greg Jones	1.0	RB	JAX	2010	1	DEN	0.9	Sunday_Morning	JAX	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	0	1	1	1
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
71947	00-0039384	Tyrone Tracy	16.0	RB	NYG	2024	17	IND	9.3	Sunday_Morning	NYG	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	0	1	1	1
71948	00-0039391	Kimani Vidal	16.0	RB	LAC	2024	17	NE	2.4	Saturday_Morning	NE	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	0	1	1	1
71955	00-0039738	Blake Corum	16.0	RB	LA	2024	17	ARI	-0.4	Saturday_Night	LA	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	0	1	1	1
71959	00-0039794	Braelon Allen	16.0	RB	NYJ	2024	17	BUF	2.6	Sunday_Morning	BUF	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	0	1	1	1
71961	00-0039798	Isaiah Davis	16.0	RB	NYJ	2024	17	BUF	5.3	Sunday_Morning	BUF	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	0	1	1	1

	player_id	player_name	game_number	position	team	season	week	opponent_team	day_slate	location	carries_lagged	rushing_yards_lagged	rushing_tds_lagged	rushing_fumbles_lagged	rushing_fumbles_lost_lagged	rushing_first_downs_lagged	rushing_epa_lagged	rushing_2pt_conversions_lagged	receptions_lagged	targets_lagged	receiving_yards_lagged	receiving_tds_lagged	receiving_fumbles_lagged	receiving_fumbles_lost_lagged	receiving_air_yards_lagged	receiving_yards_after_catch_lagged	receiving_first_downs_lagged	receiving_epa_lagged	receiving_2pt_conversions_lagged	racr_lagged	target_share_lagged	air_yards_share_lagged	wopr_lagged	special_teams_tds_lagged	fantasy_points_ppr_lagged	used_opponent_avg	used_game_number_avg	DNP_prior_season_game	never_seen	prediction	fantasy_points_ppr
24028.0	00-0031493	Jarryd Hayne	2.0	RB	SF	2015	2	PIT	Sunday_Morning	PIT	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	0	1	1	1	2.126136	0.300000
6218.0	00-0024559	D.J. Ware	6.0	RB	NYG	2011	6	BUF	Sunday_Morning	NYG	3.333333	12.166667	0.000000	0.000000	0.000000	1.166667	-0.491446	0.0	1.000000	1.333333	8.500000	0.0	0.0	0.0	4.500000	7.000000	0.333333	0.099140	0.0	0.064615	0.055585	0.014550	0.093562	0.0	3.066667	0	1	1	0	2.161232	1.900000
71813.0	00-0036875	Rhamondre Stevenson	16.0	RB	NE	2024	17	LAC	Saturday_Morning	NE	9.000000	39.000000	0.000000	1.000000	1.000000	3.000000	-2.257059	0.0	1.000000	1.000000	0.000000	0.0	0.0	0.0	0.000000	0.000000	0.000000	-0.640769	0.0	NaN	0.041667	0.000000	0.062500	0.0	2.900000	1	0	0	0	9.278624	0.100000
41527.0	00-0031665	Zach Zenner	13.0	RB	DET	2018	14	ARI	Sunday_Afternoon	ARI	2.000000	0.000000	0.000000	0.000000	0.000000	0.000000	-2.129197	0.0	0.000000	0.000000	0.000000	0.0	0.0	0.0	0.000000	0.000000	0.000000	NaN	0.0	NaN	NaN	NaN	NaN	0.0	0.000000	0	1	0	0	2.997141	11.400000
57673.0	00-0034301	Darrel Williams	3.0	RB	ARI	2022	3	LA	Sunday_Afternoon	ARI	7.000000	28.000000	0.000000	0.000000	0.000000	3.000000	0.675689	0.0	2.000000	3.000000	11.000000	0.0	0.0	0.0	2.000000	9.000000	1.000000	0.198242	0.0	5.500000	0.075000	0.006042	0.116730	0.0	5.900000	0	1	0	0	3.327356	0.700000
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
27758.0	00-0030547	Denard Robinson	14.0	RB	JAX	2015	15	ATL	Sunday_Morning	JAX	10.384615	44.769230	0.307692	0.153846	0.153846	1.769231	-1.580904	0.0	1.769231	2.384615	9.538462	0.0	0.0	0.0	-6.307693	15.461538	0.461538	-0.270148	0.0	-0.643803	0.070407	-0.045913	0.073472	0.0	8.738461	0	1	1	0	4.593217	16.700001
16221.0	00-0027029	LeSean McCoy	8.0	RB	PHI	2013	8	NYG	Sunday_Morning	PHI	23.000000	123.000000	0.000000	1.000000	0.000000	6.000000	-1.806905	0.0	3.000000	4.000000	17.000000	0.0	0.0	0.0	-10.000000	21.000000	1.000000	-0.454453	0.0	-1.700000	0.148148	-0.040323	0.193996	0.0	17.000000	1	0	0	0	16.252660	10.500000
39688.0	00-0034109	Phillip Lindsay	7.0	RB	DEN	2018	7	ARI	Thursday_Night	ARI	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	0	1	1	1	9.306647	16.600000
7753.0	00-0028087	Dion Lewis	10.0	RB	PHI	2011	11	NYG	Sunday_Night	NYG	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	0	1	1	1	3.900717	0.100000
4299.0	00-0027888	Toby Gerhart	14.0	RB	MIN	2010	15	CHI	Monday_Night	MIN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	0	1	1	1	4.443255	10.500000

	player_id	player_name	game_number	position	team	season	week	opponent_team	day_slate	location	receptions_lagged	targets_lagged	receiving_yards_lagged	receiving_tds_lagged	receiving_fumbles_lagged	receiving_fumbles_lost_lagged	receiving_air_yards_lagged	receiving_yards_after_catch_lagged	receiving_first_downs_lagged	receiving_epa_lagged	receiving_2pt_conversions_lagged	racr_lagged	target_share_lagged	air_yards_share_lagged	wopr_lagged	special_teams_tds_lagged	fantasy_points_ppr_lagged	used_opponent_avg	used_game_number_avg	DNP_prior_season_game	never_seen	prediction	fantasy_points_ppr
1163.0	00-0027761	Brody Eldridge	4.0	TE	IND	2010	4	JAX	Sunday_Afternoon	JAX	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	0	1	1	1	1.718054	2.000000
35269.0	00-0027061	Jared Cook	9.0	TE	LV	2017	9	MIA	Sunday_Night	MIA	2.888889	4.777778	35.666668	0.111111	0.111111	0.111111	46.000000	14.000000	1.777778	1.275925	0.0	0.698336	0.140991	0.153078	0.318642	0.0	6.900000	0	1	1	0	6.971167	20.600000
19493.0	00-0028042	Luke Stocker	3.0	TE	TB	2014	3	ATL	Thursday_Night	ATL	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	0	1	1	1	2.189459	3.500000
45252.0	00-0031299	Jordan Matthews	10.0	TE	PHI	2019	11	NE	Sunday_Afternoon	PHI	3.000000	4.000000	37.000000	0.000000	0.000000	0.000000	40.000000	14.000000	2.000000	-0.274862	0.0	0.925000	0.121212	0.126984	0.270707	0.0	6.700000	0	1	0	0	8.217109	1.600000
38306.0	00-0028142	Virgil Green	3.0	TE	LAC	2018	3	LA	Sunday_Afternoon	LA	3.000000	3.000000	18.000000	0.000000	0.000000	0.000000	11.000000	7.000000	1.000000	0.273514	0.0	1.636364	0.075000	0.026699	0.131189	0.0	4.800000	0	1	0	0	2.725423	1.600000
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
32474.0	00-0031610	Darren Waller	14.0	TE	BAL	2016	15	PHI	Sunday_Morning	BAL	0.666667	2.000000	6.000000	0.000000	0.000000	0.000000	28.666666	4.333333	0.333333	-0.580328	0.0	0.208709	0.054064	0.103408	0.153482	0.0	1.266667	0	1	1	0	3.810845	2.100000
50361.0	00-0033357	Taysom Hill	11.0	TE	NO	2020	12	DEN	Sunday_Afternoon	DEN	1.000000	1.000000	0.000000	0.000000	0.000000	0.000000	-4.000000	4.000000	0.000000	-1.949606	0.0	0.000000	0.026316	-0.010695	0.031987	0.0	1.000000	0	1	0	0	3.971295	17.520000
2043.0	00-0022127	Jason Witten	7.0	TE	DAL	2010	8	JAX	Sunday_Morning	DAL	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	0	1	1	1	8.921994	25.700001
51408.0	00-0034988	Kaden Smith	14.0	TE	NYG	2020	15	CLE	Sunday_Night	NYG	3.000000	3.000000	38.000000	0.000000	0.000000	0.000000	24.000000	14.000000	1.000000	1.339205	0.0	1.583333	0.103448	0.082192	0.212707	0.0	6.800000	0	1	0	0	2.926350	1.000000
68172.0	00-0038996	Tucker Kraft	4.0	TE	GB	2024	4	MIN	Sunday_Morning	GB	6.000000	6.000000	48.000000	0.000000	0.000000	0.000000	9.000000	39.000000	3.000000	1.776649	0.0	5.333333	0.176471	0.025210	0.282353	0.0	10.800000	1	0	0	0	6.748039	17.299999

	player_id	player_name	game_number	position	team	season	week	opponent_team	fantasy_points_ppr	day_slate	location	completions_lagged	attempts_lagged	passing_yards_lagged	passing_tds_lagged	interceptions_lagged	sacks_lagged	sack_yards_lagged	sack_fumbles_lagged	sack_fumbles_lost_lagged	passing_air_yards_lagged	passing_yards_after_catch_lagged	passing_first_downs_lagged	passing_epa_lagged	passing_2pt_conversions_lagged	pacr_lagged	dakota_lagged	carries_lagged	rushing_yards_lagged	rushing_tds_lagged	rushing_fumbles_lagged	rushing_fumbles_lost_lagged	rushing_first_downs_lagged	rushing_epa_lagged	rushing_2pt_conversions_lagged	receptions_lagged	targets_lagged	receiving_yards_lagged	receiving_tds_lagged	receiving_fumbles_lagged	receiving_fumbles_lost_lagged	receiving_air_yards_lagged	receiving_yards_after_catch_lagged	receiving_first_downs_lagged	receiving_epa_lagged	receiving_2pt_conversions_lagged	racr_lagged	target_share_lagged	air_yards_share_lagged	wopr_lagged	special_teams_tds_lagged	fantasy_points_ppr_lagged	used_opponent_avg	used_game_number_avg	DNP_prior_season_game	never_seen
18	00-0017915	Ricky Williams	1.0	RB	MIA	2010	1	BUF	6.200000	Sunday_Morning	BUF	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	0	1	1	1
33	00-0020516	Todd Heap	1.0	TE	BAL	2010	1	NYJ	13.200000	Monday_Afternoon	NYJ	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	0	1	1	1
36	00-0020679	Shaun Hill	1.0	QB	DET	2010	1	CHI	-0.880000	Sunday_Morning	CHI	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	0	1	1	1
46	00-0021231	David Garrard	1.0	QB	JAX	2010	1	DEN	19.799999	Sunday_Morning	JAX	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	0	1	1	1
51	00-0021429	Carson Palmer	1.0	QB	CIN	2010	1	NE	20.700001	Sunday_Morning	NE	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	0	1	1	1
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
71946	00-0039379	Jamari Thrash	16.0	WR	CLE	2024	17	MIA	0.000000	Sunday_Afternoon	CLE	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	0	1	1	1
71948	00-0039391	Kimani Vidal	16.0	RB	LAC	2024	17	NE	2.400000	Saturday_Morning	NE	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	0	1	1	1
71954	00-0039732	Bo Nix	16.0	QB	DEN	2024	17	CIN	21.860001	Saturday_Afternoon	CIN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	0	1	1	1
71962	00-0039849	Marvin Harrison	16.0	WR	ARI	2024	17	LA	15.600000	Saturday_Night	LA	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	0	1	1	1
71965	00-0039855	Jalen McMillan	16.0	WR	TB	2024	17	CAR	23.000000	Sunday_Morning	TB	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	0	1	1	1

	player_id	player_name	game_number	position	team	season	week	opponent_team	day_slate	location	completions_lagged	attempts_lagged	passing_yards_lagged	passing_tds_lagged	interceptions_lagged	sacks_lagged	sack_yards_lagged	sack_fumbles_lagged	sack_fumbles_lost_lagged	passing_air_yards_lagged	passing_yards_after_catch_lagged	passing_first_downs_lagged	passing_epa_lagged	passing_2pt_conversions_lagged	pacr_lagged	dakota_lagged	carries_lagged	rushing_yards_lagged	rushing_tds_lagged	rushing_fumbles_lagged	rushing_fumbles_lost_lagged	rushing_first_downs_lagged	rushing_epa_lagged	rushing_2pt_conversions_lagged	receptions_lagged	targets_lagged	receiving_yards_lagged	receiving_tds_lagged	receiving_fumbles_lagged	receiving_fumbles_lost_lagged	receiving_air_yards_lagged	receiving_yards_after_catch_lagged	receiving_first_downs_lagged	receiving_epa_lagged	receiving_2pt_conversions_lagged	racr_lagged	target_share_lagged	air_yards_share_lagged	wopr_lagged	special_teams_tds_lagged	fantasy_points_ppr_lagged	used_opponent_avg	used_game_number_avg	DNP_prior_season_game	never_seen	prediction	fantasy_points_ppr
38623.0	00-0029141	Alfred Morris	4.0	RB	SF	2018	4	LAC	Sunday_Afternoon	LAC	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	NaN	0.0	NaN	NaN	9.000	36.00	0.0	0.0	0.0	0.00	-0.586857	0.0	1.000	1.0	13.0	0.0	0.0	0.0	-4.000000	17.00	0.000	1.381309	0.0	-3.250000	0.040000	-0.019139	0.046603	0.0	5.90	1	0	0	0	3.168494	2.800000
7587.0	00-0025463	Jason Hill	10.0	WR	JAX	2011	11	CLE	Sunday_Morning	CLE	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	NaN	0.0	NaN	NaN	0.000	0.00	0.0	0.0	0.0	0.00	NaN	0.0	1.600	3.4	36.0	0.2	0.0	0.0	63.200001	3.60	1.400	1.337559	0.0	0.627408	0.100328	0.248784	0.324641	0.0	6.40	0	1	1	0	4.520901	7.900000
67430.0	00-0036212	Tua Tagovailoa	2.0	QB	MIA	2024	2	BUF	Thursday_Night	MIA	25.0	35.0	282.0	1.0	1.0	4.0	31.0	1.0	0.0	188.0	134.0	12.0	-3.484106	0.0	1.500000	0.063745	2.000	7.00	0.0	0.0	0.0	1.00	0.550477	0.0	0.000	0.0	0.0	0.0	0.0	0.0	0.000000	0.00	0.000	NaN	0.0	NaN	NaN	NaN	NaN	0.0	13.98	1	0	0	0	17.481883	5.500000
48234.0	00-0036145	Lawrence Cager	4.0	TE	NYJ	2020	4	DEN	Thursday_Night	NYJ	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	0	1	1	1	1.870466	0.000000
67410.0	00-0035341	Trenton Irwin	2.0	WR	CIN	2024	2	KC	Sunday_Afternoon	KC	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	NaN	0.0	NaN	NaN	0.000	0.00	0.0	0.0	0.0	0.00	NaN	0.0	1.000	2.0	4.0	0.0	0.0	0.0	25.000000	0.00	0.000	-1.106347	0.0	0.160000	0.064516	0.104167	0.169691	0.0	1.40	1	0	0	0	2.827927	1.500000
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
49723.0	00-0030578	Cordarrelle Patterson	10.0	RB	CHI	2020	10	MIN	Monday_Night	CHI	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	NaN	0.0	NaN	NaN	2.000	4.00	0.0	0.0	0.0	1.00	-0.910965	0.0	1.000	1.0	3.0	0.0	0.0	0.0	-2.000000	5.00	0.000	-0.350358	0.0	-1.500000	0.030303	-0.009756	0.038625	0.0	1.70	1	0	0	0	3.834467	12.900000
55639.0	00-0035864	Kendall Hinton	12.0	WR	DEN	2021	13	KC	Sunday_Night	KC	1.0	9.0	13.0	0.0	2.0	1.0	1.0	0.0	0.0	134.0	16.0	1.0	-11.930539	0.0	0.097015	-0.128963	2.000	7.00	0.0	0.0	0.0	0.00	-1.408619	0.0	0.000	0.0	0.0	0.0	0.0	0.0	0.000000	0.00	0.000	NaN	0.0	NaN	NaN	NaN	NaN	0.0	-2.78	0	1	1	0	1.887540	1.700000
23510.0	00-0026286	Jerome Felton	1.0	FB	BUF	2015	1	IND	Sunday_Morning	BUF	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	NaN	0.0	NaN	NaN	0.750	6.00	0.0	0.0	0.0	0.25	0.073349	0.0	0.000	0.5	0.0	0.0	0.0	0.0	-0.500000	0.00	0.000	-0.429878	0.0	0.000000	0.025016	-0.004401	0.034443	0.0	0.60	0	1	1	0	0.844864	0.000000
29828.0	00-0030496	Le'Veon Bell	6.0	RB	PIT	2016	6	MIA	Sunday_Morning	MIA	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	NaN	0.0	NaN	NaN	24.000	88.00	0.0	0.0	0.0	5.00	-5.887273	0.0	0.000	0.0	0.0	0.0	0.0	0.0	0.000000	0.00	0.000	NaN	0.0	NaN	NaN	NaN	NaN	0.0	8.80	0	1	0	0	14.846074	18.799999
57670.0	00-0034279	Equanimeous St. Brown	3.0	WR	CHI	2022	3	HOU	Sunday_Morning	CHI	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	NaN	0.0	NaN	NaN	0.375	1.75	0.0	0.0	0.0	0.25	-0.159345	0.0	0.875	1.5	9.5	0.0	0.0	0.0	13.250000	6.75	0.625	0.820310	0.0	1.568093	0.048332	0.055124	0.111085	0.0	2.00	0	1	1	0	3.077059	7.300000

	fantasy_points_ppr	never_seen	DNP_prior_season_game
count	8235.000000	8235.000000	8235.000000
mean	76.319427	0.316958	0.502008
std	82.078316	0.465257	0.431771
min	-7.280000	0.000000	0.000000
25%	10.700000	0.000000	0.000000
50%	44.799999	0.000000	0.428571
75%	120.400002	1.000000	1.000000
max	448.399994	1.000000	1.000000

	player_id	player_name	game_number	position	team	season	week	opponent_team	fantasy_points_ppr	day_slate	location	completions_lagged	attempts_lagged	passing_yards_lagged	passing_tds_lagged	interceptions_lagged	sacks_lagged	sack_yards_lagged	sack_fumbles_lagged	sack_fumbles_lost_lagged	passing_air_yards_lagged	passing_yards_after_catch_lagged	passing_first_downs_lagged	passing_epa_lagged	passing_2pt_conversions_lagged	pacr_lagged	dakota_lagged	carries_lagged	rushing_yards_lagged	rushing_tds_lagged	rushing_fumbles_lagged	rushing_fumbles_lost_lagged	rushing_first_downs_lagged	rushing_epa_lagged	rushing_2pt_conversions_lagged	receptions_lagged	targets_lagged	receiving_yards_lagged	receiving_tds_lagged	receiving_fumbles_lagged	receiving_fumbles_lost_lagged	receiving_air_yards_lagged	receiving_yards_after_catch_lagged	receiving_first_downs_lagged	receiving_epa_lagged	receiving_2pt_conversions_lagged	racr_lagged	target_share_lagged	air_yards_share_lagged	wopr_lagged	special_teams_tds_lagged	fantasy_points_ppr_lagged	used_opponent_avg	used_game_number_avg	DNP_prior_season_game	never_seen
0	00-0003292	Kerry Collins	1.0	QB	IND	2011	1	HOU	7.88	Sunday_Morning	HOU	14.0	24.0	237.0	2.0	1.0	3.0	25.0	0.0	0.0	330.0	38.0	12.0	11.256248	0.0	0.718182	0.240972	2.0	-2.0	0.0	0.0	0.0	0.0	-1.636152	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	NaN	0.0	NaN	NaN	NaN	NaN	0.0	15.280000	1	0	0	0
1	00-0003292	Kerry Collins	2.0	QB	IND	2011	2	CLE	7.54	Sunday_Morning	IND	17.0	25.0	149.0	1.0	1.0	2.0	5.0	2.0	1.0	220.0	25.0	9.0	0.543270	1.0	0.677273	0.131788	2.0	-1.0	0.0	0.0	0.0	0.0	-2.320931	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	NaN	0.0	NaN	NaN	NaN	NaN	0.0	7.860000	0	1	0	0
2	00-0003292	Kerry Collins	3.0	QB	IND	2011	3	PIT	3.72	Sunday_Night	IND	17.0	25.0	149.0	1.0	1.0	2.0	5.0	2.0	1.0	220.0	25.0	9.0	0.543270	1.0	0.677273	0.131788	2.0	-1.0	0.0	0.0	0.0	0.0	-2.320931	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	NaN	0.0	NaN	NaN	NaN	NaN	0.0	7.860000	1	0	0	0
3	00-0004541	Donald Driver	1.0	WR	GB	2011	1	NO	8.10	Thursday_Night	GB	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	NaN	0.0	NaN	NaN	0.0	0.0	0.0	0.0	0.0	0.0	NaN	0.0	5.0	7.0	30.0	1.0	0.0	0.0	54.0	2.0	4.0	1.204149	0.0	0.555556	0.233333	0.168224	0.467757	0.0	14.000000	0	1	0	0
4	00-0004541	Donald Driver	2.0	WR	GB	2011	2	CAR	2.00	Sunday_Morning	CAR	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	NaN	0.0	NaN	NaN	0.0	0.0	0.0	0.0	0.0	0.0	NaN	0.0	4.0	5.0	38.0	1.0	0.0	0.0	41.0	12.0	3.0	3.403702	0.0	0.926829	0.172414	0.146429	0.361121	0.0	13.800000	0	1	0	0
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
50498	00-0033110	Tyler Higbee	16.0	TE	LA	2024	17	ARI	2.90	Saturday_Night	LA	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	NaN	0.0	NaN	NaN	0.0	0.0	0.0	0.0	0.0	0.0	NaN	0.0	3.5	4.0	23.5	1.0	0.0	0.0	17.5	8.5	1.0	1.184914	0.0	1.379934	0.152074	0.095039	0.294638	0.0	11.849999	1	0	0	0
50499	00-0035662	Marquise Brown	15.0	WR	KC	2024	16	HOU	9.50	Saturday_Morning	KC	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	NaN	0.0	NaN	NaN	1.0	-6.0	0.0	0.0	0.0	0.0	-1.322328	0.0	2.0	5.0	18.0	0.0	0.0	0.0	78.0	2.0	1.0	-3.218636	0.0	0.230769	0.172414	0.348214	0.502371	0.0	3.200000	1	0	0	0
50500	00-0035662	Marquise Brown	16.0	WR	KC	2024	17	PIT	8.60	Wednesday_Morning	PIT	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	NaN	0.0	NaN	NaN	0.0	0.0	0.0	0.0	0.0	0.0	NaN	0.0	0.0	3.0	0.0	0.0	0.0	0.0	14.0	0.0	0.0	-2.788815	0.0	0.000000	0.130435	0.080925	0.252300	0.0	0.000000	1	0	0	0
50501	00-0036924	Michael Carter	15.0	RB	ARI	2024	16	CAR	9.80	Sunday_Morning	CAR	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	NaN	0.0	NaN	NaN	0.0	0.0	0.0	0.0	0.0	0.0	NaN	0.0	0.0	1.0	0.0	0.0	0.0	0.0	3.0	0.0	0.0	-0.758748	0.0	0.000000	0.027027	0.013575	0.050043	0.0	0.000000	0	1	0	0
50502	00-0036924	Michael Carter	16.0	RB	ARI	2024	17	LA	10.10	Saturday_Night	LA	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	NaN	0.0	NaN	NaN	4.0	19.0	0.0	0.0	0.0	1.0	0.709671	0.0	4.0	4.0	15.0	0.0	0.0	0.0	5.0	10.0	0.0	-1.802484	0.0	3.000000	0.090909	0.012788	0.145315	0.0	7.400000	1	0	0	0

	player_id	player_name	game_number	position	team	season	week	opponent_team	day_slate	location	completions_lagged	attempts_lagged	passing_yards_lagged	passing_tds_lagged	interceptions_lagged	sacks_lagged	sack_yards_lagged	sack_fumbles_lagged	sack_fumbles_lost_lagged	passing_air_yards_lagged	passing_yards_after_catch_lagged	passing_first_downs_lagged	passing_epa_lagged	passing_2pt_conversions_lagged	pacr_lagged	dakota_lagged	carries_lagged	rushing_yards_lagged	rushing_tds_lagged	rushing_fumbles_lagged	rushing_fumbles_lost_lagged	rushing_first_downs_lagged	rushing_epa_lagged	rushing_2pt_conversions_lagged	receptions_lagged	targets_lagged	receiving_yards_lagged	receiving_tds_lagged	receiving_fumbles_lagged	receiving_fumbles_lost_lagged	receiving_air_yards_lagged	receiving_yards_after_catch_lagged	receiving_first_downs_lagged	receiving_epa_lagged	receiving_2pt_conversions_lagged	racr_lagged	target_share_lagged	air_yards_share_lagged	wopr_lagged	special_teams_tds_lagged	fantasy_points_ppr_lagged	used_opponent_avg	used_game_number_avg	DNP_prior_season_game	never_seen	prediction	fantasy_points_ppr
3433.0	00-0025708	Matt Moore	5.0	QB	MIA	2011	6	NYJ	Monday_Night	NYJ	5.0	10.0	35.0	0.0	2.0	0.0	0.0	0.0	0.0	83.0	11.0	2.0	-4.838793	0.0	0.421687	0.000462	0.0	0.0	0.0	0.0	0.0	0.0	NaN	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	NaN	0.0	NaN	NaN	NaN	NaN	0.0	-2.600000	0	1	0	0	12.227106	4.160000
19787.0	00-0031062	James White	15.0	RB	NE	2016	16	NYJ	Saturday_Morning	NE	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	NaN	0.0	NaN	NaN	2.0	4.0	0.0	0.0	0.0	0.5	-0.312680	0.0	4.0	5.0	27.0	0.5	0.0	0.0	11.0	24.0	2.0	2.064640	0.0	2.866667	0.126941	0.035052	0.214949	0.0	10.100000	1	0	0	0	10.510996	13.600000
49169.0	00-0038041	Jake Ferguson	7.0	TE	DAL	2024	8	SF	Sunday_Night	SF	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	NaN	0.0	NaN	NaN	0.0	0.0	0.0	0.0	0.0	0.0	NaN	0.0	3.0	3.0	28.0	0.0	0.0	0.0	5.0	23.0	1.0	1.450288	0.0	5.600000	0.115385	0.016393	0.184552	0.0	5.800000	1	0	0	0	6.551242	8.300000
7513.0	00-0023437	Ronnie Brown	4.0	RB	LAC	2013	4	DAL	Sunday_Afternoon	LAC	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	NaN	0.0	NaN	NaN	0.0	0.0	0.0	0.0	0.0	0.0	NaN	0.0	3.0	3.0	50.0	0.0	0.0	0.0	16.0	34.0	2.0	3.993453	0.0	3.125000	0.130435	0.099379	0.265217	0.0	8.000000	0	1	0	0	3.775029	0.700000
30236.0	00-0033839	Dede Westbrook	2.0	WR	JAX	2019	2	HOU	Sunday_Morning	HOU	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	NaN	0.0	NaN	NaN	0.0	0.0	0.0	0.0	0.0	0.0	NaN	0.0	4.0	4.0	17.0	0.0	0.0	0.0	13.0	4.0	2.0	1.473923	0.0	1.307692	0.100000	0.056769	0.189738	0.0	5.700000	1	0	0	0	10.498238	1.300000
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
10543.0	00-0020337	Steve Smith	9.0	WR	BAL	2014	9	PIT	Sunday_Night	PIT	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	NaN	0.0	NaN	NaN	0.0	0.0	0.0	0.0	0.0	0.0	NaN	0.0	6.0	11.0	63.0	0.0	0.0	0.0	126.0	11.0	4.0	-1.600571	0.0	0.500000	0.343750	0.386503	0.786177	0.0	12.300000	0	1	0	0	13.510208	8.600000
37382.0	00-0034855	Baker Mayfield	2.0	QB	CLE	2021	2	HOU	Sunday_Morning	CLE	12.0	20.0	132.0	0.0	0.0	1.0	7.0	0.0	0.0	165.0	55.0	5.0	-3.269064	0.0	0.800000	0.018554	3.0	1.0	0.0	0.0	0.0	0.0	-6.779699	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	NaN	0.0	NaN	NaN	NaN	NaN	0.0	5.380000	1	0	0	0	16.614061	17.520000
29853.0	00-0032950	Carson Wentz	11.0	QB	PHI	2019	12	SEA	Sunday_Morning	PHI	20.0	28.0	236.0	1.0	0.0	3.0	22.0	0.0	0.0	159.0	125.0	11.0	8.937419	0.0	1.484277	0.134911	2.0	-2.0	0.0	0.0	0.0	0.0	-1.712713	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	NaN	0.0	NaN	NaN	NaN	NaN	0.0	13.240000	0	1	0	0	17.242601	8.940000
27450.0	00-0033906	Alvin Kamara	1.0	RB	NO	2018	1	TB	Sunday_Morning	NO	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	NaN	0.0	NaN	NaN	10.0	68.0	1.0	1.0	1.0	4.0	-0.752347	0.0	6.0	7.0	84.0	1.0	0.0	0.0	11.0	80.0	4.0	4.271715	0.0	7.636364	0.259259	0.053922	0.426634	0.0	31.200001	1	0	0	0	19.337232	43.099998
33764.0	00-0034364	Jordan Akins	8.0	TE	HOU	2020	9	JAX	Sunday_Morning	JAX	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	NaN	0.0	NaN	NaN	0.0	0.0	0.0	0.0	0.0	0.0	NaN	0.0	2.5	3.0	24.5	0.0	0.0	0.0	11.5	15.0	1.5	2.467189	0.0	4.625000	0.105296	0.035241	0.182612	0.0	4.950000	1	0	0	0	3.821753	3.900000

	player_id	player_name	season	team	position	entry_year	rookie_year
0	00-0023459	Aaron Rodgers	2025	PIT	QB	2005	2005
2	00-0024243	Marcedes Lewis	2025	CHI	TE	2006	2006
5	00-0026158	Joe Flacco	2025	CLE	QB	2008	2008
7	00-0026300	Josh Johnson	2025	WAS	QB	2008	2008
8	00-0026498	Matthew Stafford	2025	LA	QB	2009	2009
...	...	...	...	...	...	...	...
2487	00-0039919	Rome Odunze	2025	CHI	WR	2024	2024
2488	00-0039920	Malachi Corley	2025	NYJ	WR	2024	2024
2489	00-0039921	Trey Benson	2025	ARI	RB	2024	2024
2490	00-0039923	J.J. McCarthy	2025	MIN	QB	2024	2024
2500	00-0039936	Thomas Yassmin	2025	DEN	TE	2024	2024

	NFL 2025 Fixture	Unnamed: 1	Unnamed: 2	Unnamed: 3	Unnamed: 4	Unnamed: 5	Unnamed: 6	home_team	away_team	location	season	time	weekday	hour	slate	day_slate	week
0	1	1	9/4/25 20:20	Lincoln Financial Field	Philadelphia Eagles	Dallas Cowboys	9/4/25 20:20	PHI	DAL	PHI	2025	2025-09-04 20:20:00	Thursday	20	Night	Thursday_Night	1
1	2	1	9/5/25 20:00	Arena Corinthians	Los Angeles Chargers	Kansas City Chiefs	9/5/25 20:00	LAC	KC	global	2025	2025-09-05 20:00:00	Friday	20	Night	Friday_Night	1
2	3	1	9/7/25 13:00	Mercedes-Benz Stadium	Atlanta Falcons	Tampa Bay Buccaneers	9/7/25 13:00	ATL	TB	ATL	2025	2025-09-07 13:00:00	Sunday	13	Morning	Sunday_Morning	1
3	4	1	9/7/25 13:00	Huntington Bank Field	Cleveland Browns	Cincinnati Bengals	9/7/25 13:00	CLE	CIN	CLE	2025	2025-09-07 13:00:00	Sunday	13	Morning	Sunday_Morning	1
4	5	1	9/7/25 13:00	Lucas Oil Stadium	Indianapolis Colts	Miami Dolphins	9/7/25 13:00	IND	MIA	IND	2025	2025-09-07 13:00:00	Sunday	13	Morning	Sunday_Morning	1
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
267	268	18	1/5/26 0:00	MetLife Stadium	New York Giants	Dallas Cowboys	1/4/2026 19:00:00	NYG	DAL	NYG	2025	2026-01-04 19:00:00	Sunday	19	Afternoon	Sunday_Afternoon	18
268	269	18	1/5/26 0:00	Lincoln Financial Field	Philadelphia Eagles	Washington Commanders	1/4/2026 19:00:00	PHI	WAS	PHI	2025	2026-01-04 19:00:00	Sunday	19	Afternoon	Sunday_Afternoon	18
269	270	18	1/5/26 0:00	Acrisure Stadium	Pittsburgh Steelers	Baltimore Ravens	1/4/2026 19:00:00	PIT	BAL	PIT	2025	2026-01-04 19:00:00	Sunday	19	Afternoon	Sunday_Afternoon	18
270	271	18	1/5/26 0:00	Levi's Stadium	San Francisco 49ers	Seattle Seahawks	1/4/2026 19:00:00	SF	SEA	SF	2025	2026-01-04 19:00:00	Sunday	19	Afternoon	Sunday_Afternoon	18
271	272	18	1/5/26 0:00	Raymond James Stadium	Tampa Bay Buccaneers	Carolina Panthers	1/4/2026 19:00:00	TB	CAR	TB	2025	2026-01-04 19:00:00	Sunday	19	Afternoon	Sunday_Afternoon	18

Data uploading & cleaning¶

Cleaning & adding game_number¶

Lagged Stats columns¶

EDA¶

Overall¶

Null Counts¶

Summary Stats¶

Histograms¶

Correlation¶

By Position¶

QBs¶

Null Counts¶

Summary Stats¶

Histograms¶

Correlation¶

WRs¶

Null Counts¶

Summary Stats¶

Histograms¶

Correlation¶

RBs/FBs¶

Null Counts¶

Summary Stats¶

Histograms¶

Correlation¶

TEs¶

Null Counts¶

Summary Stats¶

Histograms¶

Correlation¶

CatBoost Modeling¶

QBs¶

Cross Validation¶

Exploring causes for errors¶

QBs with >=70 fantasy points in a season, seen before only¶

Cross Validation¶

WRs¶

Cross Validation¶

Exploring causes for errors¶

RBs¶

Cross Validation¶

Exploring causes for errors¶

TEs¶

Cross Validation¶

Exploring causes for errors¶

All players attempt¶

Cross Validation¶

Exploring causes for errors¶

Players with >=11 fantasy points in a season, seen before only¶

Cross Validation¶

Neural Network Modeling¶

Pre-processing dataset¶

Creating Model¶

Generating 2025 Predictions¶

2025 DF creation¶

Roster data¶

Schedule data + cleaning¶

Stats data¶

2025 DF Scorings¶

CatBoost (all players only)¶

Neural Network (all players only)¶

Pre-processing¶

Predictions¶

Top 10s by Position¶

QBs¶

WRs¶

RBs¶

TEs¶

LLM-Generated Analysis of Results¶

Function¶

CatBoost Top 10s by Position¶

QBs¶

WRs¶

RBs¶

TEs¶

Neural Network Top 10s by Position¶

QBs¶

WRs¶

RBs¶

TEs¶

	player_id	player_name	game_number	position	team	season	week	opponent_team	completions	attempts	passing_yards	passing_tds	interceptions	sacks	sack_yards	sack_fumbles	sack_fumbles_lost	passing_air_yards	passing_yards_after_catch	passing_first_downs	passing_epa	passing_2pt_conversions	pacr	dakota	carries	rushing_yards	rushing_tds	rushing_fumbles	rushing_fumbles_lost	rushing_first_downs	rushing_epa	rushing_2pt_conversions	receptions	targets	receiving_yards	receiving_tds	receiving_fumbles	receiving_fumbles_lost	receiving_air_yards	receiving_yards_after_catch	receiving_first_downs	receiving_epa	receiving_2pt_conversions	racr	target_share	air_yards_share	wopr	special_teams_tds	fantasy_points	fantasy_points_ppr	day_slate	location
66996	00-0023459	Aaron Rodgers	1.0	QB	NYJ	2024	1	SF	13	21	167.0	1	1.0	1.0	5.0	0	0	164.0	93.0	8.0	3.258283	0	1.018293	0.066574	1	-1.0	0	0.0	0.0	0.0	0.000000	0	0	0	0.0	0	0.0	0.0	0.0	0.0	0.0	NaN	0	NaN	NaN	NaN	NaN	0.0	8.580000	8.580000	Monday_Night	SF
66997	00-0026498	Matthew Stafford	1.0	QB	LA	2024	1	DET	34	49	317.0	1	1.0	2.0	13.0	0	0	290.0	171.0	15.0	4.536421	0	1.093103	0.106137	0	0.0	0	0.0	0.0	0.0	NaN	0	0	0	0.0	0	0.0	0.0	0.0	0.0	0.0	NaN	0	NaN	NaN	NaN	NaN	0.0	14.680000	14.680000	Sunday_Night	DET
66998	00-0027973	Andy Dalton	1.0	QB	CAR	2024	1	NO	0	1	0.0	0	0.0	0.0	-0.0	0	0	6.0	0.0	0.0	-0.621024	0	0.000000	NaN	0	0.0	0	0.0	0.0	0.0	NaN	0	0	0	0.0	0	0.0	0.0	0.0	0.0	0.0	NaN	0	NaN	NaN	NaN	NaN	0.0	0.000000	0.000000	Sunday_Morning	NO
66999	00-0028118	Tyrod Taylor	1.0	QB	NYJ	2024	1	SF	6	8	36.0	1	0.0	0.0	-0.0	0	0	31.0	24.0	3.0	3.210536	0	1.161290	0.242823	1	7.0	0	0.0	0.0	0.0	0.772778	0	0	0	0.0	0	0.0	0.0	0.0	0.0	0.0	NaN	0	NaN	NaN	NaN	NaN	0.0	6.140000	6.140000	Monday_Night	SF
67000	00-0029604	Kirk Cousins	1.0	QB	ATL	2024	1	PIT	16	26	155.0	1	2.0	2.0	18.0	0	0	140.0	73.0	8.0	-11.084852	0	1.107143	0.015108	1	0.0	0	0.0	0.0	0.0	-5.980317	0	0	0	0.0	0	0.0	0.0	0.0	0.0	0.0	NaN	0	NaN	NaN	NaN	NaN	0.0	6.200000	6.200000	Sunday_Morning	ATL
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
71975	00-0039915	Ladd McConkey	16.0	WR	LAC	2024	17	NE	0	0	0.0	0	0.0	0.0	0.0	0	0	0.0	0.0	0.0	NaN	0	NaN	NaN	0	0.0	0	0.0	0.0	0.0	NaN	0	8	10	94.0	2	0.0	0.0	113.0	21.0	4.0	5.864735	0	0.831858	0.250000	0.298942	0.584259	0.0	21.400000	29.400000	Saturday_Morning	NE
71976	00-0039916	Ricky Pearsall	16.0	WR	SF	2024	17	DET	0	1	0.0	0	0.0	0.0	-0.0	0	0	10.0	0.0	0.0	-0.530975	0	0.000000	NaN	2	6.0	0	0.0	0.0	0.0	-0.737734	0	8	10	141.0	1	0.0	0.0	134.0	37.0	7.0	1.708887	0	1.052239	0.277778	0.400000	0.696667	0.0	20.700001	28.700001	Monday_Night	SF
71977	00-0039917	Michael Penix	16.0	QB	ATL	2024	17	WAS	19	35	223.0	1	1.0	2.0	12.0	2	0	437.0	83.0	11.0	-0.642952	0	0.510297	0.036004	2	3.0	0	0.0	0.0	0.0	-0.584434	0	0	0	0.0	0	0.0	0.0	0.0	0.0	0.0	NaN	0	NaN	NaN	NaN	NaN	0.0	11.220000	11.220000	Sunday_Night	WAS
71978	00-0039918	Caleb Williams	16.0	QB	CHI	2024	17	SEA	16	28	122.0	0	1.0	7.0	46.0	1	0	202.0	108.0	7.0	-10.225746	0	0.603960	-0.000481	5	37.0	0	0.0	0.0	1.0	1.389207	0	0	0	0.0	0	0.0	0.0	0.0	0.0	0.0	NaN	0	NaN	NaN	NaN	NaN	0.0	6.580000	6.580000	Thursday_Night	CHI
71979	00-0039919	Rome Odunze	16.0	WR	CHI	2024	17	SEA	0	0	0.0	0	0.0	0.0	0.0	0	0	0.0	0.0	0.0	NaN	0	NaN	NaN	0	0.0	0	0.0	0.0	0.0	NaN	0	1	4	15.0	0	0.0	0.0	78.0	0.0	1.0	-1.646263	0	0.192308	0.153846	0.467066	0.557715	0.0	1.500000	2.500000	Thursday_Night	CHI

	player_name	season	completions_lagged	attempts_lagged	passing_yards_lagged	passing_tds_lagged	interceptions_lagged	sacks_lagged	sack_yards_lagged	sack_fumbles_lagged	sack_fumbles_lost_lagged	passing_air_yards_lagged	passing_yards_after_catch_lagged	passing_first_downs_lagged	passing_epa_lagged	passing_2pt_conversions_lagged	pacr_lagged	dakota_lagged	carries_lagged	rushing_yards_lagged	rushing_tds_lagged	rushing_fumbles_lagged	rushing_fumbles_lost_lagged	rushing_first_downs_lagged	rushing_epa_lagged	rushing_2pt_conversions_lagged	receptions_lagged	targets_lagged	receiving_yards_lagged	receiving_tds_lagged	receiving_fumbles_lagged	receiving_fumbles_lost_lagged	receiving_air_yards_lagged	receiving_yards_after_catch_lagged	receiving_first_downs_lagged	receiving_epa_lagged	receiving_2pt_conversions_lagged	racr_lagged	target_share_lagged	air_yards_share_lagged	wopr_lagged	special_teams_tds_lagged	fantasy_points_ppr_lagged	used_opponent_avg	used_game_number_avg	DNP_prior_season_game	never_seen	prediction
0	A.J. Barner	2025	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.00	0.00	0.00	0.0	0.0	0.0	0.000000	0.0	32.346154	44.576923	252.461548	1.461538	0.0	0.0	158.846161	145.576920	16.653847	1.569527	0.0	33.860535	1.389639	0.876495	2.698004	0.0	66.361542	7	9	2	0	43.415744
1	A.J. Brown	2025	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.00	0.00	0.00	0.0	0.0	0.0	0.000000	0.0	81.807692	114.423077	1318.000000	9.576923	0.0	0.0	1383.692261	413.807678	63.846153	85.014107	0.0	15.834282	5.399290	8.319300	13.922444	0.0	271.069214	7	9	2	0	234.540318
2	A.J. Dillon	2025	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.00	0.00	0.00	0.0	0.0	0.0	0.000000	0.0	0.000000	0.000000	0.000000	0.000000	0.0	0.0	0.000000	0.000000	0.000000	0.000000	0.0	0.000000	0.000000	0.000000	0.000000	0.0	0.000000	0	16	16	16	42.832999
3	A.T. Perry	2025	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.00	0.00	0.00	0.0	0.0	0.0	0.000000	0.0	0.000000	0.000000	0.000000	0.000000	0.0	0.0	0.000000	0.000000	0.000000	0.000000	0.0	0.000000	0.000000	0.000000	0.000000	0.0	0.000000	0	16	16	16	31.218986
4	Aaron Jones	2025	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	259.00	1304.00	9.00	1.0	1.0	57.5	0.581283	0.0	43.000000	52.000000	399.000000	1.000000	0.0	0.0	79.500000	358.500000	19.500000	20.500946	0.0	28.504810	1.788492	0.276181	2.876065	0.0	271.299988	8	8	0	0	198.693876
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
752	Zamir White	2025	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	146.75	420.25	1.75	6.5	6.5	10.0	-101.826530	0.0	10.500000	14.000000	28.500000	0.000000	0.0	0.0	-37.750000	52.500000	1.500000	-9.526148	0.0	-8.409091	0.636607	-0.294303	0.748898	0.0	52.875000	4	12	6	0	51.299977
753	Zavier Scott	2025	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.00	0.00	0.00	0.0	0.0	0.0	0.000000	0.0	0.000000	0.000000	0.000000	0.000000	0.0	0.0	0.000000	0.000000	0.000000	0.000000	0.0	0.000000	0.000000	0.000000	0.000000	0.0	0.000000	0	16	16	16	30.798918
754	Zay Flowers	2025	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	7.50	51.50	0.00	0.0	0.0	1.0	3.050954	0.0	71.500000	117.000000	1097.500000	3.500000	0.0	0.0	1394.000000	467.000000	50.500000	52.228844	0.0	12.753320	4.297002	5.083618	10.004036	0.0	207.399994	8	8	0	0	187.448584
755	Zay Jones	2025	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.00	0.00	0.00	0.0	0.0	0.0	0.000000	0.0	13.200000	19.800000	138.399994	0.000000	0.0	0.0	212.199997	48.200001	10.400001	-3.913983	0.0	8.364285	0.571616	1.168927	1.675672	0.0	27.040001	3	13	9	0	37.474360
756	Zonovan Knight	2025	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.00	0.00	0.00	0.0	0.0	0.0	0.000000	0.0	0.000000	0.000000	0.000000	0.000000	0.0	0.0	0.000000	0.000000	0.000000	0.000000	0.0	0.000000	0.000000	0.000000	0.000000	0.0	0.000000	0	16	16	16	35.673772

	CatBoost QB	CatBoost 2025 Fantasy Points	CatBoost QB 2024 Fantasy Points	Neural Network QB	Neural Network 2025 Fantasy Points	Neural Network 2024 Fantasy Points	2024 QB	2024 QB Fantasy Points
Rank
1	Lamar Jackson	320.725420	407.399994	Jalen Hurts	367.313629	315.119995	Lamar Jackson	407.399994
2	Josh Allen	316.677465	372.339996	Lamar Jackson	348.037872	407.399994	Josh Allen	372.339996
3	Kyler Murray	314.890762	269.359985	Josh Allen	338.890747	372.339996	Joe Burrow	359.839996
4	Justin Herbert	311.388292	257.360016	Patrick Mahomes	332.876617	283.019989	Jayden Daniels	351.600006
5	Jalen Hurts	310.555803	315.119995	Kyler Murray	329.036957	269.359985	Baker Mayfield	344.160004
6	Patrick Mahomes	306.552962	283.019989	Joe Burrow	326.817993	359.839996	Jared Goff	315.220001
7	Joe Burrow	294.815278	359.839996	Jordan Love	320.277252	231.099991	Jalen Hurts	315.119995
8	Tua Tagovailoa	293.090861	181.580002	Brock Purdy	315.933533	266.860016	Sam Darnold	300.320007
9	Dak Prescott	281.102271	116.520004	Aaron Rodgers	315.812531	231.619995	Bo Nix	283.660004
10	Jared Goff	278.546773	315.220001	Carson Wentz	308.521606	0.800000	Patrick Mahomes	283.019989

	CatBoost WR	CatBoost 2025 Fantasy Points	CatBoost WR 2024 Fantasy Points	Neural Network WR	Neural Network 2025 Fantasy Points	Neural Network 2024 Fantasy Points	2024 WR	2024 WR Fantasy Points
Rank
1	Ja'Marr Chase	261.747430	377.399994	Rashee Rice	378.356079	64.900002	Ja'Marr Chase	377.399994
2	Justin Jefferson	247.197554	309.080017	Amon-Ra St. Brown	337.724121	302.480011	Justin Jefferson	309.080017
3	CeeDee Lamb	242.578282	263.399994	Justin Jefferson	337.223236	309.080017	Amon-Ra St. Brown	302.480011
4	Mike Evans	238.848725	222.500000	Puka Nacua	333.984161	206.600006	Brian Thomas	266.700012
5	Terry McLaurin	238.347535	247.599991	CeeDee Lamb	318.362915	263.399994	CeeDee Lamb	263.399994
6	A.J. Brown	234.540318	216.899994	Ja'Marr Chase	301.477539	377.399994	Malik Nabers	256.200012
7	Amon-Ra St. Brown	234.106467	302.480011	Davante Adams	297.966003	220.500000	Jaxon Smith-Njigba	247.699997
8	Tyreek Hill	228.847087	214.199997	DK Metcalf	292.935120	176.899994	Terry McLaurin	247.599991
9	Chris Godwin	227.732478	137.800003	Deebo Samuel	283.767853	155.600006	Garrett Wilson	242.800003
10	Courtland Sutton	222.819952	219.500000	Chris Olave	281.625824	76.700005	Drake London	240.100006

	CatBoost RB	CatBoost 2025 Fantasy Points	CatBoost RB 2024 Fantasy Points	Neural Network RB	Neural Network 2025 Fantasy Points	Neural Network 2024 Fantasy Points	2024 RB	2024 RB Fantasy Points
Rank
1	Alvin Kamara	286.006225	265.299988	Christian McCaffrey	357.399963	47.800003	Saquon Barkley	355.300018
2	Christian McCaffrey	253.567084	47.800003	Bijan Robinson	354.477997	310.399994	Bijan Robinson	310.399994
3	Jonathan Taylor	233.095463	221.699997	Jahmyr Gibbs	341.096924	308.899994	Jahmyr Gibbs	308.899994
4	David Montgomery	229.837736	221.720001	Alvin Kamara	300.254761	265.299988	Derrick Henry	306.299988
5	James Conner	229.664879	253.800003	Breece Hall	293.621521	222.699997	Devon Achane	281.799988
6	Kareem Hunt	216.403447	155.399994	Kyren Williams	292.006683	272.100006	Josh Jacobs	281.500000
7	Saquon Barkley	214.697060	355.300018	Jonathan Taylor	275.383698	221.699997	Kyren Williams	272.100006
8	Derrick Henry	213.331433	306.299988	Bucky Irving	273.310028	228.000000	Alvin Kamara	265.299988
9	Breece Hall	212.944621	222.699997	Kenneth Walker	264.775787	181.199997	James Cook	257.899994
10	Rachaad White	211.670768	199.600006	David Montgomery	260.497589	221.720001	Chase Brown	255.000000

	CatBoost TE	CatBoost 2025 Fantasy Points	CatBoost TE 2024 Fantasy Points	Neural Network TE	Neural Network 2025 Fantasy Points	Neural Network 2024 Fantasy Points	2024 TE	2024 TE Fantasy Points
Rank
1	Travis Kelce	217.626596	195.399994	Travis Kelce	251.315414	195.399994	Brock Bowers	247.699997
2	George Kittle	187.373434	231.900009	George Kittle	246.789642	231.900009	George Kittle	231.900009
3	Pat Freiermuth	173.546287	145.800003	Mark Andrews	241.647049	173.399994	Trey McBride	224.300003
4	Mark Andrews	167.471897	173.399994	Brock Bowers	225.114166	247.699997	Jonnu Smith	203.699997
5	Sam LaPorta	164.449828	161.300003	Erick All	223.334000	35.799999	Travis Kelce	195.399994
6	T.J. Hockenson	163.930067	83.599998	Trey McBride	220.372559	224.300003	Mark Andrews	173.399994
7	Dallas Goedert	152.992022	94.099998	Sam LaPorta	215.898071	161.300003	Zach Ertz	162.000000
8	Brock Bowers	147.044759	247.699997	T.J. Hockenson	214.280060	83.599998	Sam LaPorta	161.300003
9	Trey McBride	142.400545	224.300003	Jake Ferguson	210.288605	101.199997	Tucker Kraft	155.699997
10	Dalton Kincaid	137.434067	100.800003	Zach Ertz	195.989990	162.000000	David Njoku	148.500000

	player_id	player_name	season	gemini_analysis
0	00-0035659	Terry McLaurin	2025	Rating: 72/100. The projection of 238.35 fanta...
1	00-0036963	Amon-Ra St. Brown	2025	Rating: 72/100. The 234-point projection for A...
2	00-0031408	Mike Evans	2025	Rating: 73/100. Mike Evans is predicted to sco...
3	00-0033921	Chris Godwin	2025	Rating: 78/100. Chris Godwin is projected to s...
4	00-0034348	Courtland Sutton	2025	Rating: 70/100. Courtland Sutton is projected ...
5	00-0035676	A.J. Brown	2025	Rating: 68/100. A.J. Brown is predicted to sco...
6	00-0036322	Justin Jefferson	2025	Rating: 79/100. A total of 247.20 predicted fa...
7	00-0033040	Tyreek Hill	2025	Rating: 72/100. Tyreek Hill is projected for 2...
8	00-0036358	CeeDee Lamb	2025	Rating: 78/100. CeeDee Lamb is projected to sc...
9	00-0036900	Ja'Marr Chase	2025	Rating: 67/100. Ja'Marr Chase is projected to ...

	player_id	player_name	season	gemini_analysis
0	00-0038120	Breece Hall	2025	Rating: 68/100. Breece Hall's projection of 21...
1	00-0033280	Christian McCaffrey	2025	Rating: 62/100. The projection of 253.57 fanta...
2	00-0033923	Kareem Hunt	2025	Rating: 68/100. Kareem Hunt's predicted 216.40...
3	00-0032764	Derrick Henry	2025	Rating: 72/100. Derrick Henry's projection of ...
4	00-0033906	Alvin Kamara	2025	Rating: 68/100. Alvin Kamara is projected 286....
5	00-0035685	David Montgomery	2025	Rating: 72/100. The projected 229.84 fantasy p...
6	00-0037256	Rachaad White	2025	Rating: 68/100. Rachaad White is projected for...
7	00-0034844	Saquon Barkley	2025	Rating: 68/100. Saquon Barkley is projected 21...
8	00-0036223	Jonathan Taylor	2025	Rating: 78/100. Jonathan Taylor is projected f...
9	00-0033553	James Conner	2025	Rating: 78/100. The projection of 229.66 PPR p...

	player_id	player_name	season	gemini_analysis
0	00-0036894	Pat Freiermuth	2025	Rating: 68/100. Pat Freiermuth's projected tot...
1	00-0033288	George Kittle	2025	Rating: 72/100. The predicted 187.37 fantasy p...
2	00-0030506	Travis Kelce	2025	Rating: 72/100. Travis Kelce is projected to s...
3	00-0034753	Mark Andrews	2025	Rating: 72/100. The projection of 167.47 PPR p...
4	00-0039338	Brock Bowers	2025	Rating: 72/100. Brock Bowers is projected to s...
5	00-0039065	Sam LaPorta	2025	Rating: 72/100. The predicted total of 164.45 ...
6	00-0038933	Dalton Kincaid	2025	Rating: 70/100. The prediction of 137.43 fanta...
7	00-0034351	Dallas Goedert	2025	Rating: 68/100. Dallas Goedert's predicted 152...
8	00-0035229	T.J. Hockenson	2025	Rating: 72/100. T.J. Hockenson is projected to...
9	00-0037744	Trey McBride	2025	Rating: 67/100. The predicted total of 142.40 ...

	player_id	player_name	season	gemini_analysis
0	00-0035640	DK Metcalf	2025	Rating: 70/100. DK Metcalf is projected to sco...
1	00-0035719	Deebo Samuel	2025	Rating: 79/100. Deebo Samuel is projected for ...
2	00-0031381	Davante Adams	2025	Rating: 79/100. Davante Adams is projected for...
3	00-0039075	Puka Nacua	2025	Rating: 72/100. Puka Nacua is projected for 33...
4	00-0039067	Rashee Rice	2025	Rating: 65/100. Rashee Rice's projected total ...
5	00-0037239	Chris Olave	2025	Rating: 68/100. Chris Olave is projected for 2...
6	00-0036963	Amon-Ra St. Brown	2025	Rating: 70/100. The predicted total of 337.72 ...
7	00-0036322	Justin Jefferson	2025	Rating: 72/100. A projection of 337.22 fantasy...
8	00-0036358	CeeDee Lamb	2025	Rating: 72/100. CeeDee Lamb's projection of 31...
9	00-0036900	Ja'Marr Chase	2025	Rating: 72/100. The predicted season-long tota...